A review of the properties of pyrite and the implications for

Transcription

Technical Report
TR-13-19
A review of the properties of pyrite
and the implications for corrosion
of the copper canister
Fraser King, Integrity Corrosion Consulting Limited
December 2013
Svensk Kärnbränslehantering AB
Swedish Nuclear Fuel
and Waste Management Co
Box 250, SE-101 24 Stockholm Phone +46 8 459 84 00
ISSN 1404-0344
Tänd ett lager:
SKB TR-13-19
P, R eller TR.
ID 1377791
A review of the properties of pyrite
and the implications for corrosion
of the copper canister
Fraser King, Integrity Corrosion Consulting Limited
December 2013
Keywords: Pyrite, Dissolution, Corrosion, Canister, Copper, Sulphide, Polysulphide,
Bentonite.
This report concerns a study which was conducted for SKB. The conclusions
and viewpoints presented in the report are those of the author. SKB may draw
modified conclusions, based on additional literature sources and/or expert opinions.
A pdf version of this document can be downloaded from www.skb.se.
Figures 2-1, 4-2 and 4-3 reprinted with permission of the copyright owner, the American Chemical Society.
Figures 2-2, 2-3, 3-1, 3-2, 6-1 and 6-3 reprinted with permission of the copyright owner, Elsevier.
Figure 3-3, copyright owner G. Delécaut.
Figure 3-4 reprinted with permission of the copyright owner, the Materials Research Society.
Summary
A review of the properties of pyrite impurities in bentonite clay has been carried out based on infor
mation available in the literature. The main focus of the review is the extent to which pyrite dissolu
tion during the long-term anaerobic period could support corrosion of the canister, most likely as
the result of the release of dissolved sulphide or polysulphide species. In addition, the oxidative dis
solution of pyrite has also been considered since processes occurring during the initial oxic period,
and potentially continuing into the anaerobic phase, might affect the behaviour of pyrite during the
anaerobic phase and its overall impact on canister corrosion.
Various aspects of the properties and behaviour of pyrite have been reviewed, including:
• the structure, composition, and electronic and thermodynamic properties of pyrite,
• surface characterisation and reactivity of pyrite,
• the relationship between pyrite and other iron sulphides,
• the characterisation of pyrite present in bentonite clays,
• attempts at mathematical modelling of the behaviour of pyrite in the repository,
• the aqueous speciation of sulphide and polysulphides,
• the solubility of pyrite and other iron sulphides,
• the dissolution behaviour of pyrite, including oxidative, reductive, and chemical (congruent)
dissolution mechanisms,
• electrochemical studies of pyrite, and
• the expected behaviour of pyrite in the repository and the implications for the corrosion of
the canister.
Based on the information available, it is considered unlikely that pyrite will be a significant source
of sulphide or oxidants for corrosion of the canister. Instead, the predominant role of pyrite is likely
to be the consumption of a fraction of the initially trapped atmospheric oxygen in the repository. The
solubility of pyrite is so low that very little sulphide will be released during the long-term anaerobic
phase even over repository timescales.
SKB TR-13-193
Contents
1Introduction7
2
Properties of pyrite and associated iron sulphides9
2.1Pyrite
9
2.1.1Structure
9
2.1.2Composition
9
2.1.3Semi-conductivity
10
2.1.4 Thermodynamic properties
10
2.1.5 Surface characterisation
10
2.1.6 Surface reactivity – gas phase studies
11
2.1.7 Interaction with metal ions
11
2.2 Formation of pyrite and its relationship to other iron sulphides
12
3
Pyrite in bentonite15
3.1Background
15
3.2 Characterisation and quantification of pyrite in bentonite and sedimentary
host rocks
15
3.2.1 Pyrite in bentonite
15
3.2.2 Pyrite in sedimentary host rocks
18
3.3 Experimental studies of the behaviour of pyrite in bentonite
18
3.4 Modelling studies of the behaviour of pyrite under repository conditions
19
3.4.1 Canister corrosion
19
3.4.2 Oxygen consumption
20
3.4.3 Impact on processes in the buffer and host rock under anaerobic
conditions20
4
4.1
4.2
4.3
Aqueous chemistry of sulphides and polysulphides21
Speciation of dissolved sulphide
21
Speciation of dissolved polysulphides
22
Dissolved iron-sulphur species
23
5
Solubility of pyrite and other iron sulphides25
5.1 Mackinawite and greigite
25
5.2Pyrite
27
6
6.1
6.2
6.3
6.4
Pyrite dissolution and electrochemistry31
Oxidative dissolution
31
Reductive dissolution
34
Chemical dissolution
36
Electrochemistry of pyrite
36
7
7.1
7.2
7.3
Implications for repository performance39
Initial state of pyrite in bentonite
39
Evolution of pyrite behaviour in repository
39
7.2.1 In the buffer
39
7.2.2 In the backfill
41
Consequences for corrosion of the canister
41
8
Summary and conclusions45
References47
SKB TR-13-195
1Introduction
Pyrite (FeS2p) is the most common sulphide mineral in surface environments on the Earth and is
found in sedimentary, igneous, and metamorphic geological formations, as well as in deep sea vents
(Rickard and Luther 2007). (The p subscript is used here to distinguish pyrite from the dimorph mar
casite FeS2m). The oxidation of pyrite leads to the phenomenon of acid rock drainage which can have
significant environmental impacts (Vaughan 2005). Pyrite is also the principal source of inorganic
sulphur in coal (Zhao et al. 2005), the combustion of which results in air pollution and associated
environmental effects.
Pyrite is an impurity mineral in bentonite and other clays proposed for use as sealing materials for the
planned KBS-3 repository for the disposal of spent nuclear fuel in Sweden (Karnland 2010, Pusch
2001). The pyrite may have a number of effects on the evolution of the repository and the behaviour
of the canister. First, oxidation of pyrite during the early warm, aerobic phase will consume a por
tion of the initially trapped atmospheric O2 in the buffer and backfill materials (King et al. 2010,
Puigdomenech et al. 2001). Second, during the long-term anaerobic phase, equilibria involving
pyrite may control the redox potential Eh in the repository (Puigdomenech et al. 2001). Third, pyrite
may be a source of sulphide ions HS– for corrosion of the canister (SKB 2010a). For the SR-Site
safety assessment, the extent of corrosion due to pyrite in the buffer is conservatively estimated
based on a mass-balance approach, despite the belief that the availability of sulphide will be limited
by the low solubility of pyrite. Corrosion due to the release of HS– from pyrite in the backfill is
treated using a mass-transport argument (SKB 2010a).
Figure 1‑1 shows a schematic of different phases in the formation and alteration of pyrite in the
context of the use of bentonite in the KBS-3 repository. Pyrite is formed either by precipitation
from a high-temperature melt or by the precipitation of Fe(II) and HS– and the subsequent ageing
of a sequence of increasingly thermodynamically stable iron sulphides. The formation of bentonites
also involves exposure to extreme temperatures and to aqueous environments, being the product of
volcanic ash which is subsequently hydrothermally altered (Christidis and Huff 2009). Thus, either
process is a feasible pathway for the formation of the pyrite found in bentonite deposits. Depending
upon the environmental conditions to which the bentonite deposit is exposed, alteration of the pyrite
is possible over geological timescales prior to exploitation of the deposit. Mining and processing
of the raw bentonite may also expose the pyrite impurities to further (oxidation) alteration. Once
emplaced in the repository, the pyrite will be exposed to a warm, aerobic phase and may undergo
further alteration. Finally, once anaerobic conditions have become established in the repository,
pyrite may dissolve and release sulphide ions. The focus of the current review is the potential for
pyrite to act as a source of sulphide for canister corrosion during this latter long-term anaerobic
phase, although it is important to understand that the original pyrite phase may have been exten
sively altered during the intervening periods.
Pyrite is an Fe(II) polysulphide with a cubic crystal structure and exhibits semiconducting properties
(Rickard and Luther 2007). Thus, the dissolution behaviour of pyrite can be discussed in electro
chemical terms, with both oxidative and reductive dissolution mechanisms possible (Figure 1‑2).
Under oxidising conditions in the presence of O2 or Fe(III), pyrite will dissolve oxidatively with
the formation of Fe(II) and SO42– (note, it is the polysulphide species that is oxidised rather than
the metal species). Under reducing conditions, FeS2p dissolves with the release of HS– (i.e., the
polysulphide species S22– with a nominal average oxidation state of –I is reduced to a sulphide with
oxidation state –II). Under natural conditions, the reduction of FeS2p must be accompanied by the
oxidation of a suitable reducing species. At intermediate Eh values, pyrite may dissolve chemically
with the release of Fe(II) and S22–, with the latter species undergoing disproportionation in the aque
ous phase. In terms of the various stages illustrated in Figure 1‑2, the focus of the current report is
on the chemical and reductive dissolution processes.
SKB TR-13-197
Exposure of altered
pyrite to warm aerobic
phase immediately aer
repository closure
Alteraon of pyrite
over millenia in
bentonite deposit
Dissoluon of
altered pyrite and
release of HS- and
other potenally
corrosive sulphur
species
Alteraon of pyrite
during mining and
processing of bentonite
Formaon of pyrite during
bentonite genesis by precipitaon
from aqueous phase or via
high-temperature melt
Current
Figure 1-1. Lifecycle of pyrite from formation in bentonites to alteration in the repository.
S(-I) → S(VI)
Potential
as S(-II)
as Fe(II) and S22-
Fe(III) → Fe(II)
O2 → OH-
Figure 1-2. Schematic of the electrochemical behaviour of pyrite.
Various aspects of the dissolution behaviour of pyrite are reviewed here. First, the properties of pyrite
are described, including the crystal structure, electronic, and thermodynamic properties. The relation
ship of pyrite to other iron sulphides is also discussed, since transformation of pyrite to other moresoluble phases could represent a route by which higher concentrations of dissolved sulphide could
occur. Next, the nature and extent of pyrite in bentonite clay are considered, although the available
information is limited because of the difficulties associated with characterising minor phases. In the
next two sections, the aqueous chemistry of dissolved sulphur species and the solubility of solid iron
sulphides are discussed. The speciation of aqueous sulphur species is relevant as pyrite may dissolve
with the release of sulphate, polysulphide, or sulphide, depending upon the redox potential. The
solubilities of mackinawite and greigite as well as that of pyrite are considered here. The dissolution
behaviour of pyrite is discussed in the next section, with a brief description of the oxidative dissolu
tion mechanism, a more-detailed review of the available information on reductive dissolution, and a
summary of the chemical dissolution behaviour. In addition, the electrochemical behaviour of FeS2p
is also discussed. Finally, implications of the foregoing information for the dissolution behaviour of
pyrite in the repository under oxic and anaerobic conditions and the implications for corrosion of the
canister are summarised.
8
SKB TR-13-19
2
Properties of pyrite and associated iron sulphides
2.1Pyrite
2.1.1Structure
Pyrite is an Fe(II) polysulphide with the nominal composition FeS2p. Pyrite has a cubic NaCl-type
structure with Fe(II) at the cube corners and face centres and S22– at the cube centre and at the mid
points on the cube edges. Figure 2‑1 shows the crystal structure of pyrite, with the polysulphide ions
orientated as shown (Rickard and Luther 2007). Because of the arrangement of the polysulphide,
pyrite exhibits a lower degree of symmetry than NaCl and consequent chirality, which results in
differences in reactivity of different crystal planes.
2.1.2Composition
Although close to stoichiometric, naturally occurring pyrite exhibits a wide range of minor and trace
elements (Abraitis et al. 2004). Trace elements, defined here as species with a maximum concentra
tion of 0.1 wt.% (1,000 wppm) include Te, Sn, Se, Ru, Pt, Pd, Cd, and Ag. Minor elements found
with maximum concentrations of 1 wt.% (10,000 wppm) include Zn, Tl, Sb, Pb, Ni, Hg, and Au.
The impurity elements found at highest concentration in natural pyrites (in the range 1–10 wt%) are
Mo, Cu, Co, and As. The latter two impurity elements are interesting because of their impact on the
semi-conducting natural of pyrite, with As also of particular interest because of the environmental
issues associated with acid rock drainage. These minor and trace elements can be present as either
substitutions in the pyrite lattice or as inclusions.
Figure 2-1. Crystal structure of pyrite FeS2p (Rickard and Luther 2007). The major axes of symmetry are
also shown in the figure.
SKB TR-13-199
2.1.3Semi-conductivity
Pyrite exhibits both n-type and p-type semi-conductivity and has a band gap of ~0.9 eV (Vaughan
2005). Various conduction and valence band structures have been proposed (Murphy and Strongin
2009). The nature of the impurities and the manner in which the pyrite was formed tends to deter
mine the type of semi-conductivity and the electrical resistivity (Savage et al. 2008). Pyrite formed
in sedimentary deposits at lower temperatures is primarily a p-type semi-conductor in the absence of
Cu, as are As-containing pyrites or pyrites with a S:Fe ratio > 2. Pyrite formed at higher temperatures
tends to be n-type (in the absence of As), as are pyrites containing Co as an impurity and pyrites with
S:Fe < 2. Although the resistivity of pyrites varies over a wide range, pyrites exhibiting p–type semiconductivity tend to be more resistive (reported resistivities of 0.5–35 Ω·cm) than n-type pyrites
(resistivities 0–0.64 Ω·cm) (Abraitis et al. 2004, Savage et al. 2008).
2.1.4 Thermodynamic properties
Various publications list thermodynamic properties for FeS2p although many of them are references
to other studies, with apparently few original sources. Table 2‑1 lists a number of the reported standard
free energy of formation (∆Gf0) values along with a description of their source. The majority of values
that have been reported are between –159.5 kJ/mol and –160.23 kJ/mol, although only two original
experimental sources have been found (Grønvold and Westrum 1976, Toulmin and Barton 1964).
The value of –166.9 kJ/mol listed in the US National Bureau of Standards compilation (Wagman
et al. 1982) may be in error due to an error in the ∆Gf0 value for Fe2+ (Rickard and Luther 2007).
For the current purposes, the ∆Gf0 value of –160.229 kJ/mol from Robie et al. (1978) is used for
the solubility calculations in Section 5.2.
2.1.5 Surface characterisation
Characterisation of the nature of the pyrite surface is important for understanding a range of impor
tant industrial and natural processes involving pyrite, including the oxidation behaviour, flotation of
pyrite during mineral and coal processing, and the interaction with metal ion contaminants. Studies
have been performed with naturally weathered samples, crushed and ground pyrite, as well as freshly
fractured (“pristine”) pyrite surfaces.
Pyrite does not exhibit the preferential cleavage along the [001] crystal plane typical of minerals with
a rock salt structure, but instead fractures conchoidally, i.e., along a curved fracture plane (Nesbitt
et al. 1998). Pyrite has both Fe-S and S-S bonds, with the weaker S-S cleaving preferentially. The
resulting exposed S atoms with a formal oxidation state of –I relax via an electron transfer from Fe(II),
resulting in surface S(–II) and Fe(III) species, rather than by electron transfer between S atoms
resulting in S(0) and S(–II) species (Nesbitt et al. 1998). The surface S(–II) represents an impurity
defect site, in addition to the structural defects on the surface (kinks, Fe vacancies, steps), all of
which impact the reactivity of the pristine cleaved surface.
Of more interest here is the nature of the naturally weathered pyrite surface. A range of iron surface
species have been proposed for the weathered (oxidised) surface, including: ferrous carbonate (siderite)
or sulphate, with small amounts of goethite (α‑FeOOH) (Descostes et al. 2001); α‑FeOOH, hematite
α‑Fe2O3, or magnetite Fe3O4 (Cai et al. 2009); ferric sulphate Fe2(SO4)3, ferric hydroxide Fe(OH)3,
and FeO (de Donato et al. 1993); and hydrated Fe(III) (Joeckel et al. 2005). The range of proposed
surface sulphur species is, if anything, more diverse, and includes: polysulphide Sn2–, sulphur-oxyanions,
such as thiosulphate S2O32– (Descostes et al. 2001); molecular sulphur S8 (Toniazzo et al. 1999); sul
phite SO32–, sulphate SO42–, S2O32–, and S8 (Cai et al. 2009), and polysulphide S62– and S8 (de Donato
et al. 1993). In addition, a range of other mineral phases has also been reported, including: gypsum
CaSO4·2H2O and alumino-sulphate minerals (Joeckel et al. 2005); and a range of Al and Si alteration
products, such as alunite KAl3(SO4)2(OH)6, jarosite KFe3(SO4)2(OH)6, and gibbsite Al(OH)3 formed
by the precipitation of the constituents of aluminosilicate minerals dissolved in the low pH condi
tions associated with the oxidative dissolution of pyrite (Langmuir 1997).
It is apparent, therefore, that a range of species can be expected in the crust of alteration products
on weathered pyrite surfaces. Most importantly for the current discussion is that these alteration
products can include polysulphide species, which could then dissolve and dissociate into sulphide
ions HS–, and Fe(III) species, which could promote oxidative dissolution of pyrite once anaerobic
conditions have become established in the repository.
10
SKB TR-13-19
Table 2-1. Summary of reported free energy of formation values for pyrite.
Reference
Source
–166.9
Wagman et al. 1982
Not given
–160.1*
Singh and Pourbaix 1997
Reference to another CEBELCOR report
DGf0
(kJ/mol)
Uncertainty
(kJ/mol)
–160.229
1.715
Robie et al. 1978
Reference to Toulmin and Barton 1964
–160.1
1.7
Robie and Hemingway 1995
References to Toulmin and Barton 1964 and Chase 1985
–160.060
Chase 1998
Not given
–160.1
Kaye & Laby 2005
Not given
–166.9
Kaye and Laby 1986
Not given
Grønvold and Westrum 1976
Based on original heat capacity measurements by authors
–159.5*
Toulmin and Barton 1964
Based on original experimental study by the authors
–160.23
Benning et al. 2000
Reference to Robie et al. 1978
–159.5
Rickard and Luther 2007
Inferred value from K*1sp,pyrite value of 10–14.2
–160.1
Hummel et al. 2002
Reference to Robie and Hemingway 1995
–160.2
Langmuir 1997
Reference to Robie et al. 1978
–162.2*
2.1
* Converted from cal/mol based on 1 cal = 4.184 J.
2.1.6 Surface reactivity – gas phase studies
Different crystal planes exhibit different reactivity. By necessity, the reactivity of different crystal
planes has been studied under well-controlled environmental conditions, often under vapour or gas
phase conditions under partial vacuum where the effects of different surfaces are more apparent
(Murphy and Strongin 2009). The most commonly studied surface is the [100] plane. In the gas
phase, H2O adsorbs via an Fe-O interaction but, most importantly, the water does not dissociate
and can be desorbed again as H2O. Similarly, the exposure of a [100] pyrite surface to H2S results
in adsorption of the gaseous species (and subsequent desorption) without any indication of reaction
with the pyrite surface (at temperatures below 500 K). On the assumption that these observations can
be extrapolated to aqueous conditions, these results would imply that the presence of H2O and/or HS–
under anaerobic conditions in the repository will not cause any alteration of the (pristine) pyrite surface.
In contrast to the behaviour of H2O and H2S, O2 and especially O2/H2O gas mixtures (Elsetinow et al.
2000, Guevremont et al. 1998a, b) cause oxidation of pyrite (Murphy and Strongin 2009). Oxygen
alone results in the formation of Fe-O bonds and the oxidation of S2–, indicating a degree of electron
transfer. In the presence of O2, H2O does dissociatively adsorb (unlike in the absence of O2), with the
oxygen from the H2O molecule becoming incorporated into the SO42– formed when pyrite is oxidised
by O2 or Fe(III).
In the presence of (humid) air under atmospheric conditions, pyrite surfaces oxidise (Murphy and
Strongin 2009). The [111] plane is more reactive than the [100] plane, possibly because of more Fe
atoms on the [111] plane or because of enhanced dissociative adsorption of water on the [111] plane.
Surface Fe species play a key role in the oxidation of pyrite by O2, with electron transfer between
adsorbed Fe(II) and Fe(III) species promoting the cathodic reduction of O2, the first electron transfer
to which is considered to be the rate-determining step. Blocking or chelating Fe(III) surface sites
suppresses pyrite oxidation. Surface sulphur species are also oxidised, with surface S oxidised to SO42–
and sub-surface S oxidised to SO32– (Murphy and Strongin 2009). A notable feature of these surface
studies is that the oxidation is heterogeneous across the surface with patches of Fe(II)/Fe(III) species.
Because of the absence of a preferred cleavage plane for pyrite, crushed or ground pyrite samples
exhibit a range of reactivity due to the presence of different crystal planes of different reactivity. The
most common surface plane in crushed samples is the [100] plane, with additional amounts of [021],
[111], and [110] planes (Murphy and Strongin 2009).
2.1.7 Interaction with metal ions
The interaction of pyrite with metal cations has been extensively studied, because of the importance
of the activation of pyrite during flotation, the co-precipitation and recovery of precious metals such
as gold, and the sequestration of metal contaminants (Murphy and Strongin 2009). These studies
SKB TR-13-1911
have included the study of the adsorption of Cu(II) by pyrite, primarily due to the effect of Cu in
enhancing pyrite separation by flotation (von Oertzen et al. 2007), but of interest here because of
the presence of Cu(II) in the bentonite in the transitional period between aerobic and anaerobic
phases in the evolution of the repository environment.
There is near-unanimous agreement among different authors that the adsorption of Cu(II) is accom
panied by the reduction of the adsorbate to Cu(I) and the oxidation of the adsorbent (FeS2p). Where
there is no agreement is on the nature of the oxidised species. Using X‑ray photoelectron spectroscopy
(XPS) with and without the use of synchrotron radiation, von Oertzen et al. (2007) showed that Cu(I)
was bound to both monosulphide (Fe-S2–) and polysulphide (Fe‑S22–) sites on the pyrite surface. By
analogy with other studies, the authors proposed that the overall adsorption reaction involved the
oxidation of pyrite to sulphate, according to:
FeS2p + 3xCu2+ + 4xH2O → Cu3xFeS2–x + xSO42– + 8xH+(2-1)
Laajalehto et al. (1999) found that the Cu:Fe ratio on the surface increased with decreasing Eh in acid
solution (but not at pH 9) and suggested the formation of a chalcopyrite CuFeS2-type surface layer.
This would infer that the reduction of Cu(II) to Cu(I) is accompanied by the oxidation of Fe(II) to
Fe(III), reflecting the oxidation states of these species in chalcopyrite. Weisener and Gerson (2000)
proposed the formation of an oxidised polysulphide which they denoted S22–ox, but it is not clear what
such a species is and whether sulphide is released to solution as a consequence. Using a combina
tion of XPS, X‑ray absorption near-edge structure (XANES), and extended X‑ray absorption fine
structure (EXAFS), Naveau et al. (2006) were unable to find any oxidation products on the surface
of pyrite on which cupric ions had been reduced to Cu(I) and concluded that the reductive adsorption
process was accompanied by the oxidation of S22–, the products of which had gone into solution.
2.2
Formation of pyrite and its relationship to other iron sulphides
Pyrite is just one of a large number of iron sulphides found in natural systems. Table 2‑2 lists some
of the more common minerals found along with some of their selected properties (Rickard and
Luther 2007). The inter-relationship between these sulphides is important as any process that may
lead to the conversion of pyrite to a less thermodynamically stable iron sulphide could lead to an
increase in the concentration of dissolved HS–.
Figure 2-2 shows the inter-relationship between the various iron sulphides, with known processes
indicated by the solid lines and inferred processes indicated by the dashed lines. This figure should
not be interpreted as indicating that the conversion from one phase to another occurs via a solid-state
process. Although this is possible in some cases, conversion from one sulphide to another is also
possible by the dissolution and re-precipitation of Fe(II), and/or dissolved sulphide and polysulphide
species. In general, the progression from left to right in the figure corresponds to the formation of
increasingly thermodynamically stable solids. There is no indication in the literature that any of these
transformations is reversible, i.e., once the more stable iron sulphide is formed it is not subsequently
converted to a less-stable solid (Butler and Rickard 2000).
Table 2-2. Summary of solid iron sulphides and their properties.
Mineral
Composition
Structure
Properties
Mackinawite
FeSm
Tetragonal
Metastable, principal species formed by precipitation of Fe(II)
in aqueous solution
Cubic FeS
FeSc
Troilite
FeSt
Pyrrhotite
Fe1–xS
Various
Stable, non-stochiometric iron sulphide with x < 0.2
Greigite
Fe3S4g
Inverse spinel
Metastable mixed Fe(II)/Fe(III) inverse thiospinel
Pyrite
FeS2p
Cubic
Marcasite
FeS2m
12
Unstable, formed prior to FeSm
Stoichiometric end-member of the pyrrhotite group
Stable Fe(II) disulphide
Metastable Fe(II) disulphide
SKB TR-13-19
In aqueous solution, mackinawite FeSm is the predominantly observed solid phase when Fe or Fe(II)
first reacts with HS– or H2S, particularly in environments with excess Fe(II). The kinetics of FeSm
precipitation are fast, with the mackinawite typically formed as nanoparticulate material (Rickard
and Luther 2007). So-called “amorphous FeS” is sometimes reported and can be viewed as a precur
sor of FeSm which forms from the ageing of the amorphous material. Mackinawite is metastable
with respect to pyrite but is not a necessary precursor to pyrite formation. Because of a common
cubic close packed sulphur lattice, however, mackinawite can convert to greigite Fe3S4g via a solid
state process involving rearrangement of Fe atoms. Greigite exhibits an inverse spinel structure,
with Fe(II) in tetrahedral sites and Fe(II) and Fe(III) sharing octahedral lattice positions. Chemically,
the formation of Fe3S4g from FeSm involves only the oxidation of Fe(II) to Fe(III), since both are
sulphides. In contrast, since FeS2p is an Fe(II) polysulphide, the formation of pyrite from mackina
wite would formally involve the oxidation of S(–II) to S22– but no change in oxidation state of the
Fe. The alternative route for the formation of FeS2p is via the sulphidation of pyrrhotite (or troilite).
(Figure 2‑2).
A
B
C
D
E
F
G
H
I
J
K
L
M,N
O
P
Q
formation from Fe2+ or Fe and HS- or H2S at a pH of 4–5 (T < 523 K)
ageing of cubic FeS to FeSm
solid-state conversion of FeSm to Fe3S4g
inferred transformation of greigite to marcasite
inferred sulphidation of greigite at pH > 5 (T < 523 K)
solid state transformation
sulphidation of Fe by H2S at a pH of 3–10 (T < 523 K)
precipitation of dissolved Fe2+ and HSageing of amorphous precipitate
solid state transformation, kinetics not established
transformation of greigite to pyrrhotite
precipitation of Fe2+ by HS- at pH 4–6 (wide range of temperatures to above 573 K)
reaction of Fe2+ and HS- or H2S at a pH 4–5 (or speculative transformation of proposed
wurtzite-FeS structure) to form either (M) hexagonal pyrrhotite (T > 413 K) or (N)
troilite (T < 413 K)
pyrrhotite transformation to marcasite
pyrite formed by sulphidation of pyrrhotite (this reaction is rapid above 573 K)
troilite–hexagonal pyrrhotite reversible phase transition
Figure 2-2. Inter-relationship between various iron sulphide solid phases (Vaughan 2005). The solid and
dashed lines indicate known and speculative processes, respectively.
SKB TR-13-1913
The discussion above focuses on the formation of pyrite via dissolution-precipitation processes
in the aqueous phase, which is the most likely pathway for pyrite formed in sedimentary deposits
(Schoonen 2004). However, nucleation of pyrite from solution requires a considerable degree of
supersaturation of the solution (Rickard and Luther 2007). An alternative route is via the cooling of
high-temperature molten systems (Vaughan 2005). Since bentonite is formed from the weathering of
volcanic ash, the pyrite may have formed as a high-temperature melt and then been dispersed with
the ash during an igneous event.
The relative thermodynamic stabilities of the different iron sulphides can be represented on E-pH
diagrams (Rickard and Luther 2007, Singh and Pourbaix 1997, Vaughan 2005). Figure 2‑3 shows
the potential-pH stability diagram for the Fe-S-H2O system at 25°C, illustrating the relative thermo
dynamic stabilities of the different iron sulphides and oxides. At high dissolved sulphur activity
(aStot = 0.1, Figure 2‑3A), pyrite is predicted to be stable over a wide range of pH under relatively
reducing conditions. The extent of the pyrite stability field diminishes with decreasing S activity
(aStot = 10–6, Figure 2‑3B), with Fe3O4 predicted to be more stable at pH > 9. More interestingly,
under moderately alkaline conditions and at the H2O/H2 equilibrium potential and slightly higher,
pyrrhotite FeS is more stable than FeS2p. This would imply that, in an environment that becomes
increasingly anaerobic, pyrite could be reduced to pyrrhotite (involving the reduction of S22– to S2–),
although there are no reports of such a process occurring in nature and the reaction could be kineti
cally hindered (see Section 6.2).
Figure 2-3. Potential-pH diagrams for the Fe-S-H2O system at 25°C at 1 atm total pressure and for total
dissolved sulphur activities of (A) 0.1 and (B) 10–6 (Vaughan 2005).
14
SKB TR-13-19
3
Pyrite in bentonite
3.1Background
Pyrite is present as one of a number of accessory minerals in bentonite, a list that includes quartz,
feldspars, gypsum, and various iron oxides/hydroxides (Karnland 2010). Although only present in
minor amounts, pyrite influences the evolution of the repository environment in a number of ways:
• Consumption of the initially trapped O2 leading to the onset of anoxic conditions (Grandia et al. 2006,
Lazo et al. 2003, Puigdomenech et al. 2000, 2001, Sidborn and Neretnieks 2003, Wersin et al. 1994).
• Acting as a potential source of sulphide for corrosion of the canister (King et al. 2011a, b,
SKB 2010a).
• Controlling the redox conditions during the long-term anoxic phase (Arcos et al. 2008, Pusch 2003).
• Immobilising radionuclides released from the spent fuel after failure of the canister (Bruggeman
et al. 2007, Delécaut 2004, Descostes et al. 2010, Scott et al. 2007).
Of these different processes, the one of current interest is the release of HS– ions and the subsequent
corrosion of the canisters. In order that the buffer material not significantly impair the barrier function
of the canister, one of the design premises for the buffer is that (SKB 2010b):
The sulphide content should not exceed 0.5 wt-% of the total mass, corresponding
to approximately 1% of pyrite.
The total sulphur content (including the sulphide) should not exceed 1 wt-%.
Along with a limit on the organic carbon content of the buffer, this design premise has been estab
lished to minimise the extent of canister corrosion based on the assumption that pyrite dissolution
under anaerobic conditions is a source of sulphide ions.
In this Chapter, the available information on the occurrence and nature of pyrite in bentonite and
sedimentary host rocks is reviewed. Experimental and modelling studies of the effect of pyrite on
various processes related to the evolution of the repository environment and corrosion of the canister
are also considered. The review has been limited to studies conducted for various national nuclear
waste management programmes.
3.2
Characterisation and quantification of pyrite in bentonite
and sedimentary host rocks
3.2.1 Pyrite in bentonite
The pyrite contents of a number of sources of bentonite have been reported. The pyrite content is esti
mated based on either an analysis of the sulphur content of the clay or on quantitative X-ray diffraction
(XRD) analyses (Karnland 2010). The former method involves analysis of the total S content based on
combustion at 1,200°C and analysis for sulphate based on a similar combustion at 800°C, the sulphide
content then being given by the difference in the two values (Karnland 2010). Quantitative XRD analy
sis involves the fitting of the XRD patterns using commercial software. Based on the variability of the
measured pyrite contents on duplicate samples reported by Karnland (2010), the reproducibility of both
measurements may be as high as ±30–50%. Karnland et al. (2006) suggest an accuracy for XRD of
±1 wt.% suggesting that, in the case of pyrite, the XRD analyses should be considered semi-quantita
tive and that the combustion method provides a more reliable quantitative analysis.
Table 3‑1 summarises the results of analyses of the pyrite content of various samples of bentonite from
a number of sources reported by Karnland et al. (2006). Analyses are given for both combustion and
XRD methods.
SKB TR-13-1915
It is apparent from the data in Table 3‑1 that the pyrite content of bentonite varies from deposit to
deposit and even within a given deposit. The bentonite samples from the Czech Republic come from
four different deposits, referred to as the Dnesice (Dn), Rokle (Ro), Zelena-Skalna (Sk), and Strance
(St) deposits. Based on the results of the combustion analyses, these bentonites seem to contain little
pyrite, which is confirmed by separate analyses of bentonite from the Rokle deposit which indicates
an FeS2p content of 0.00044 wt.% (Kolaříková et al. 2010). The three Danish and five Indian sam
ples, taken from different locations within the respective deposits (Karnland et al. 2006), also show
very little pyrite based on the combustion analyses.
In contrast to these bentonite sources, the samples from deposits in Germany (Friedland), Greece
(Deponit CA-N) and the USA (MX-80) contain of the order of 0.5–1.0 wt.% pyrite (based on the
combustion analyses). The pyrite contents shown in Table 3-1 are for replicate samples from the
same batch of each material, indicating both sample-to-sample variability and the reproducibility
of the measurement technique. Karnland et al. (2006) also report the results of XRD analyses of six
batches of MX-80 bentonite received over a 20‑year period, which showed the same mean pyrite
content as the reference WyR1 material with a slightly larger standard deviation.
Table 3-1. Summary of pyrite analyses for various bentonites (Karnland et al. 2006).
Country of origin
Combustion
XRD
Bentonite sample
Pyrite content (wt.%)
Bentonite sample
Pyrite content (wt.%)
Czech Republic
RoR1
SkR1
StR1
StR1
0.00
0.06
0.00
0.00
DnR1
RoR1
SkR1
StR1
0.8
1.1
0.6
0.9
Denmark
HoR1
RöR1
ÖlR1
0.00
0.00
0.96
HoR1
RöR1
ÖlR1
1.1
1.0
1.1
Germany
FrR1
FrR1
FrR1
FrR1
FrR1
Mean FrR1
Stand. Dev. FrR1
0.62
0.71
0.69
0.65
0.74
0.68
0.04
FrR1a
FrR1b
FrR1c
Mean FrR1
Stand. Dev. FrR1
1.4
0.9
1.7
1.3
0.4
Greece
MiR1
MiR1
MiR1
MiR1
MiR1
MiR1
Mean MiR1
Stand. Dev. MiR1
0.41
0.54
0.47
0.51
0.57
0.38
0.48
0.07
MiR1a
MiR1b
MiR1c
Mean MiR1
Stand. Dev. MiR1
1.2
1.2
1.0
1.1
0.1
India
Ku36R1
Ku37R1
Ku38R1
Ku39R1
Ku40R1
0.00
0.00
0.00
0.00
0.09
Ku36R1
Ku37R1
Ku38R1
Ku39R1
Ku40R1
0.9
0.5
0.5
1.1
0.3
USA
WyR1
WyR1
WyR1
WyR1
WyR1
Mean WyR1
s.d. WyR1
0.24
0.30
0.08
0.26
0.31
0.24
0.09
WyR1a
WyR1b
WyR1c
Mean WyR1
s.d. WyR1
0.5
0.6
0.6
0.6
0.1
16
SKB TR-13-19
The pyrite content of other types of bentonite that have or are being considered for buffer material in
other national nuclear waste programmes include: 0.5–1.0 wt.% in the Japanese Kunigel‑V1 (Ochs
et al. 2004, Ohkubo et al. 2008), 0.7 wt.% in German Volclay KWK (Van Loon et al. 2007), and
0.02 wt.% in the ENRESA-supplied bentonite for the FEBEX experiment (Fernández et al. 2004).
Although there have been a significant number of analyses of the amount of FeS2p in bentonite, there
is little information concerning the nature of the pyrite. The fact that pyrite can be detected by XRD
indicates the presence of crystalline material, but there is little information about the size or shape
of the FeS2p particles. Figure 3‑1 shows a photograph of a Wyoming bentonite thin section that was
subjected to synchrotron-based XRD (S‑µXRD, Lange et al. 2010) and which contains a small pyrite
particle (spot T1.2) that was specifically identified by the S-µXRD technique.
Fukushi et al. (2010) characterised unprocessed bentonite deposits from the Zao region in Japan.
Of the three samples examined, pyrite was observed in only one, in the form of greenish veins sur
rounding bentonite particles. The pyrite was present as well-defined, faceted crystals with a size of
20–25 µm (Figure 3‑2), and was believed to have been formed during exposure to an aqueous phase
at a temperature < 100°C, implying formation of FeS2p via a dissolution-precipitation process involv
ing various iron sulphides rather than via a high-temperature melt. Several of the bentonite deposits
examined by Karnland et al. (2006) are also thought to have been formed by exposure to marine or
aqueous environments, again implying pyrite formation via a dissolution-precipitation process.
Figure 3-1. Photograph of bentonite thin section used for synchrotron micro X-ray diffraction (Lange et al.
2010). Spot T1.2 was identified as pyrite.
Figure 3-2. Euhedral pyrite particle in Japanese unprocessed bentonite deposit (Fukushi et al. 2010).
The scale bar is 5 µm long.
SKB TR-13-1917
3.2.2 Pyrite in sedimentary host rocks
Pyrite is also a significant component of several sedimentary host rock formations considered
for nuclear waste repositories in various countries. Because of their potential significance on the
performance of the respective repositories, these pyrite inclusions have been characterised in some
detail. For example, the pyrite content of Opalinus Clay is ~1 wt.% (Nagra 2002). Techer et al. (2009)
describe the presence of framboids and discrete crystals of pyrite, which showed evidence for oxi
dation during long-term experiments in the Mont Terri URL and during storage in the laboratory.
(Framboids are spherical agglomerations of pyrite particles with an appearance similar to a raspberry,
the French word for which is framboise). Upon oxidation, these particles became covered by a layer
of calcium sulphate.
The Belgian Boom Clay contains up to 9 wt.% pyrite (Delécaut 2004, Zhang et al. 2008). The pyrite
takes a number of forms, including large concreted nodules of several cm in dimension, to smaller
aggregates, and framboidal particles and discrete crystals (Figure 3‑3, Delécaut 2004).
3.3
Experimental studies of the behaviour of pyrite in bentonite
Although pyrite is expected to have a number of impacts on the repository environment, there have
been few experimental studies of these processes using bentonite samples. Similar to the observation
of the oxidation of pyrite in Opalinus Clay reported by Techer et al. (2009), Kolaříková et al. (2010)
observed the formation of gypsum (CaSO4·2H2O) when Rokle bentonite was exposed to hydrothermal
conditions at temperatures up to 100°C for a period of 45 months. The appearance of gypsum was
interpreted as being the result of the oxidation of pyrite in the bentonite or present in graphite added
to the experiment to improve the heat transfer characteristics. In contrast Melamed and Pitkänen (1994,
1996) observed a decrease in the amount of gypsum following exposure of bentonite to water at
75°C for periods of up to 36 months, believed to be due to dissolution and ion-exchange of Ca2+ by
the MX‑80 bentonite. More interestingly, the authors observed the development of concentric rings
of reddish-brown corrosion products on pyrite particles in the bentonite, identified by XRD to con
tain goethite and siderite (Figure 3‑4).
Lazo et al. (2003) studied the redox properties of bentonite-NaCl solution slurries, specifically the
consumption of dissolved O2. Two types of bentonite were used, MX-80 containing 0.3 wt.% pyrite
and Montigel containing very little. Despite the differences in pyrite content of the two clays, the
rate of O2 consumption was similar, leading the authors to conclude that the oxidation of pyrite may
not be a major contributor to the consumption of the initially trapped oxygen in a repository.
Figure 3-3. Framboidal and discrete particulate pyrite in Boom Clay (Delécaut 2004). The scale bar is 10 µm.
18
SKB TR-13-19
Figure 3-4. Oxidised pyrite particle in MX-80 bentonite (after Melamed and Pitkänen 1994). The diameter
of the oxidised particle is approximately 800 µm.
3.4
Modelling studies of the behaviour of pyrite under
repository conditions
In contrast to the small number of experimental studies of the behaviour of pyrite in bentonite, there
have been a large number of modelling studies related to the effect of pyrite on canister corrosion, the
consumption of O2, and changes in the mineralogy and other properties of the bentonite buffer.
3.4.1 Canister corrosion
In the SR-Site safety assessment, it is assumed that pyrite dissolution can lead to the corrosion of the canister
(SKB 2010a). For pyrite in the buffer, the extent of corrosion is estimated based on both mass-balance and
mass-transport arguments. For the mass-balance argument, it is conservatively assumed that each pyrite mole
cule will dissolve to form two sulphide molecules which in turn will support the corrosion of four Cu atoms
4Cu + FeS2p + 4H2O + 2e– → 2Cu2S + Fe2+ + 2H2 + 4OH–(3-1)
Whilst this reaction is mass balanced, it is not electron balanced since pyrite is a polysulphide rather than
a sulphide. Regardless, the maximum depth of corrosion to the sides of the canister based on this approach
is predicted to be 0.1 mm and 0.9 mm for MX-80 and Ibeco-RWC bentonites, respectively (SKB 2010a).
(The pyrite contents of MX-80 and Ibeco-RWC bentonites are assumed to be 0.07 wt.% and 0.5 wt.%,
respectively). If the limited solubility of pyrite and the low diffusivity of HS– are taken into account, only
the pyrite within 2 cm of the canister surface will contribute to corrosion over a 106 timescale, resulting in
1 µm of corrosion for the reference values of the solubility and diffusivity (SKB 2010a).
King et al. (2011a, b) have developed a detailed reactive-transport model for predicting the extent of cor
rosion of the canister due to HS– from various sources, including pyrite dissolution, microbial sulphate
reduction, and the ground water itself. In the Copper Sulphide Model (CSM), pyrite is assumed to both
react with O2 during the aerobic phase, according to
FeS2p + 3.75O2 + 3.5H2O → Fe(OH)3 + 2SO42– + 4H+(3-2)
and to dissolve under anaerobic conditions to release sulphide, according to
FeS2p + 0.75H2O → Fe(II) + 1.5HS– + 0.25S2O32–(3-3)
The assumption that the dissolution of pyrite results in the formation of both HS– and thiosulphate ions
is based on the study of the disproportionation of polysulphide in aqueous solutions (Licht and Davis
1997). Simulations using the CSM predict long-term corrosion rates due to all sources of sulphide of the
order of < 1 nm/year, of the same order of magnitude as those predicted in for the SR-Site safety assess
ment (SKB 2010a). More importantly, of the different sources of sulphide, pyrite dissolution is predicted
to account for < 5% of the total amount of corrosion (King et al. 2011b).
SKB TR-13-1919
3.4.2 Oxygen consumption
In common with other nuclear waste management programmes, SKB has considered the impact
of the oxidation of pyrite on the consumption of the initially trapped O2 in the repository (Grandia
et al. 2006, Puigdomenech et al. 2000, 2001, Sidborn and Neretnieks 2003, Wersin et al. 1994). The
stoichiometry of the reaction between FeS2p and O2 is typically given by Reaction (3-2). The reaction
has been assumed to be both abiotic in nature (e.g., Grandia et al. 2006) or microbially mediated (e.g.,
Sidborn and Neretnieks 2003). Predictions of the time to consume the initially trapped O2 range from a
few months (Grandia et al. 2006) to as long as 300 years (Wersin et al. 1994), although these predic
tions are inconsistent with the experimental observations of Lazo et al. (2003) discussed above who
observed no apparent effect of pyrite content of bentonite on the rate of O2 consumption.
Zhang et al. (2008) describe a numerical model to predict the evolution of Boom Clay subject to heating,
irradiation, and microbial activity. Pyrite dissolution and precipitation were described by the reaction
FeS2p + H2O + 3.5O2 = Fe2+ + 2SO42– + 2H+(3-4)
although there is no evidence to suggest that this reaction is reversible and the fact that the oxidation of
S22– to SO42– (and the reverse process) involves the transfer of an average of seven electrons per S atom
makes the reversibility of this reaction extremely unlikely. Zhang et al. (2008) suggested an equilibrium
constant for Reaction (3-4) of 10224.64 at 16.5°C, which implies that the reaction would be well over to the
right-hand side, regardless of any questions regarding the kinetics of the reduction process. Nevertheless,
little net pyrite dissolution or precipitation was predicted to occur in Boom Clay subject to either heat
ing alone or heating and irradiation. In order to predict the precipitation of pyrite that was observed
in the CERBERUS experiment at the HADES facility in Mol, Belgium, Zhang et al. (2008) extended
their model to include a microbial component. The activity of both iron-reducing (IRB) and sulphatereducing bacteria (SRB) was incorporated into the model, along with an additional process for the
dissolution/precipitation of pyrite
Fe2+ + 2HS– + 0.5O2 = FeS2p + H2O(3-5)
although the basis for this reaction is unclear since pyrite is not formed under oxic conditions. With the
inclusion of this reaction and the IRB and SRB microbial activity, however, Zhang et al. (2008) were
able to predict the precipitation of FeS2p, in apparent agreement with experimental observations.
3.4.3 Impact on processes in the buffer and host rock under anaerobic conditions
There have been a number of attempts to include the effect of pyrite on the evolution of the bentonite
and far-field under anaerobic conditions. Arcos et al. (2008) suggested that the redox potential in the
bentonite in a KBS-3 repository is close to that in the granitic ground water, which in turn appears to
be close to equilibrium with pyrite and siderite.
In the French nuclear waste management programme, the reactive-transport code KIRMAT has been
used to predict the alteration of bentonite (Marty et al. 2010), transport and reactions in the bentonite
(Montes-H et al. 2005a, b), and cation exchange (Montes-H et al. 2005c). The dissolution rate of a
mineral m (vsdm) in the KIRMAT code is treated using a generalised expression of the form
Q
pH eff n 
vsdm = k dm
Sm a H+ 1 − m
K
m


 (3-6)

eff
where kpH
dm is a pH-dependent dissolution rate constant, S m is the effective surface area of the mineral,
n
+
aH+ is the activity of H ions in solution and the value of the exponent n is typically positive in acid
solution, zero at neutral pH and negative in basic solution, and Qm and Km are the ion activity product
and the equilibrium constant, respectively. The reaction for the dissolution of pyrite is not given, but
an equilibrium constant of 10–67.89 at 100°C is quoted (Marty et al. 2010, Montes‑H et al. 2005a, b, c).
Furthermore, the value given for the dissolution rate constant is (i) independent of pH, (ii) used by
Marty et al. (2010) to predict the rate of both pyrite dissolution and precipitation, and (iii) appears to
have been taken from a literature study of the oxidative dissolution of pyrite, despite the modelling
being conducted under supposed anaerobic conditions. In a separate study using the CRUNCH reactivetransport code, Bildstein et al. (2006) have also used an apparent oxidative dissolution rate constant to
describe the dissolution and precipitation of pyrite under anaerobic conditions when modelling ironbentonite interactions in a French repository.
20
SKB TR-13-19
4
Aqueous chemistry of sulphides and polysulphides
4.1
Speciation of dissolved sulphide
The dissociation of dissolved H2S in aqueous solution occurs in two stages
H 2S ⇔ HS− + H +
K1 =
{HS− }{H + }
(4-1a)
{H 2S}
K2 =
{S2 − }{H + }
(4-1b)
{HS− }
and
HS− ⇔ S2 − + H +
where the curly brackets denote activity. The value of K1 is well characterised and at infinite dilution
has a value of 10–6.98±0.03 at 25°C and 1 bar pressure (Rickard and Luther 2007). The value of K2 is
less well characterised since the formation of polysulphides at alkaline pH complicates its measure
ment. The best estimate for pK2 is > 18 (Rickard and Luther 2007). Figure 4‑1 shows the predicted
distribution of sulphide species based on Reactions (4-1a) and (4-1b) with pK1 and pK2 values of –6.98
and –18, respectively, for a total dissolved sulphide concentration of 10–5 mol·dm–3, similar to that in the
ground water at Forsmark. It is important to note that S2– is virtually non-existent in aqueous solution.
The temperature dependence of pK1 at infinite dilution is given by (Rickard and Luther 2007)
pK1 = –98.080 + 5765.4/T + 15.0455 ln T
(4-2)
where T is the temperature in K. Rickard and Luther (2007) also provide the dependence of pK1 on
salinity (for seawater).
10–4
Concentraon (mol dm–3)
10–6
10–8
10–10
10–12
10–14
H2S
10–16
HS–
10
–18
S2–
10–20
2
4
6
8
10
12
pH
Figure 4-1. Concentrations of H2S, HS , and S for a total dissolved sulphide concentration of 10–5 mol·dm–3.
–
2–
SKB TR-13-1921
4.2
Speciation of dissolved polysulphides
Polysulphide ions consist of a chain of sulphur atoms attached to a sulphide ion in the general form Sn2–
(Rickard and Luther 2007). Polysulphides with n values of 2–8 have been characterised by Kamyshny
et al. (2004). Polysulphide species dissociate in aqueous solution in an analogous fashion to the sulphides
H 2Sn ⇔ HS−n + H +
K1 ' =
{HS−n }{H + }
(4-3a)
{H 2Sn }
and
HS−n ⇔ S2n − + H +
K2 ' =
{S2n − }{H + }
(4-3b)
{HS−n }
In (alkaline) aqueous solution in the presence of excess S(0), the predominant polysulphide species
are (in order of decreasing concentration): S52–, S62–, S42–, S72–, S32–, S82–, and S22– (Kamyshny et al. 2004,
Rickard and Luther 2007). The species HS2– is the most-predominant protonated species over a wide
range of pH and is the predominant polysulphide species at pH ≤ 7. Figure 4‑2 shows the distribution
of sulphide and polysulphide species as a function of pH (Rickard and Luther 2007). The HS– ion
is the predominant dissolved species up to ~pH 9.5, with the various un-protonated polysulphides
becoming increasingly dominant at higher pH. At pH 7 the ratio of the total polysulphide concentra
tion to the concentration of HS– is ~0.01.
Polysulphide ions will also participate in redox reactions involving various oxidised sulphur species,
such as sulphate
S22– + H2O = 1.75HS– + 0.25SO42– + 0.25H+(4-4)
Figure 4-2. Distribution of sulphide and polysulphide species as a function of pH in the presence of excess
S(0) for a total dissolved sulphur concentration of 5·10–2 mol·dm–3 (Rickard and Luther 2007).
22
SKB TR-13-19
Figure 4‑3 shows the distribution of sulphide, polysulphide, and sulphate as a function of redox poten
tial for various pH values and a total dissolved sulphur concentration of 10–3 mol·dm–3 (Rickard and
Luther 2007). It is interesting to note that, in the absence of elemental sulphur, the bisulphide species
HS2– and S22– are predicted to be more stable than the higher polysulphides. Furthermore, at pH 8
(i.e., close to the pH of bentonite pore water), the ratio of the concentration of HS– to that of either
HS2– or S22– is > 1010 at the Eh corresponding to the H2/H2O equilibrium line, suggesting that HS–
will dominate the speciation of dissolved sulphide and polysulphide under anaerobic conditions.
4.3
Dissolved iron-sulphur species
Sulphide will also form complexes with Fe(II) ions in solution. A range of possible dissolved Fe‑S
complexes and clusters is described by Rickard and Luther (2007), including: Fe(HS)+, Fe(HS)2,
Fe(HS)3–, Fe2(HS)3+, Fe3(HS)5+, FeS4, Fe2S42+, FeS5, Fe2S52+, FeS, and Fe2S2. The relative abundance
of these species depends on pH and the relative Fe2+ and HS– concentrations, but they tend to be
more predominant in HS– rich environments. Further discussion of the stability and distribution
of these species can be found in Rickard and Luther (2007).
Figure 4-3. Distribution of sulphide and polysulphide species and sulphate as a function of redox potential
at various pH for a total dissolved sulphur concentration of 10–3 mol·dm–3 (Rickard and Luther 2007).
SKB TR-13-1923
5
Solubility of pyrite and other iron sulphides
As a general rule, iron sulphides are relatively insoluble, especially under the redox conditions and
pH expected in the repository. For this reason, there are few experimental studies of the solubility of
the more stable iron sulphides in the literature, and none at all for pyrite. As an alternative to measur
ing the solubility experimentally, the solubility product can be estimated from the free energy change
for an assumed dissolution process. The validity of this latter approach has been demonstrated for
mackinawite and greigite, for which experimental solubility measurements are available. The solubility
data for mackinawite and greigite are discussed first before extending the thermodynamic approach
to the prediction of the solubility of pyrite.
5.1
Mackinawite and greigite
The solubility of FeSm has been measured by a number of workers, including recently by Rickard
(2006). Earlier studies have been reviewed by Davison (1991). Rickard (2006) found that the solu
bility of FeSm (expressed in terms of the concentration of dissolved Fe(II)) exhibited a pH-dependent
region at pH < 6–8 (depending on sulphide concentration) and a pH-independent region at higher
pH. In the “low-pH” region, dissolution of mackinawite could be described by
FeSm + 2H+ → Fe2+ + H2S(5-1)
In the “high-pH” region, the stable aqueous species was considered to be dissolved FeS0. The overall
solubility could be expressed as (Rickard 2006)
log ∑ [Fe(II)] = log K 0 + log K *sp,1 − log{H 2S} − 2pH (5-2)
where K0 is the intrinsic solubility of FeSm (log K0 = –5.7) and K*sp,1 is the solubility product for
Reaction (5-1)
K *sp,1 =
{Fe 2 + }{H 2S}
(5-3)
{H + }2
and has a value of 103.5±0.25 at 23°C.
Figure 5‑1 shows a comparison of some of the measured total sulphide concentrations of Rickard
(2006) and predicted H2S or HS– concentrations (or, more strictly, activities) calculated on the basis
of the equilibrium expression for Reaction (5-1) (for pH < 7) and the corresponding reaction involv
ing HS– (for pH ≥ 7), namely:
FeSm + H+ = Fe2+ + HS–(5-4)
The free energies of formation used for the various species are given in Table 5‑1 and the free energy
change for various reactions are given in Table 5‑2.
For this comparison, data were selected from the study of Rickard (2006) with dissolved Fe(II)
concentrations close to 10–4, 10–6, or 10–8 mol·dm–3 and the corresponding measured total sulphide
concentrations were then plotted on the figure. Good agreement is found between measured and
predicted sulphide concentrations suggesting that the solubility of mackinawite can be reasonably
predicted based on the available thermodynamic data. As noted above, however, Rickard (2006)
reported that the solubility of mackinawite was independent of pH in alkaline solution (pH > 8) so
that the thermodynamic prediction is likely to be less reliable at elevated pH.
SKB TR-13-1925
4
10–8 Fe2+
2
Rickard (2006)
log10 [S(-II)]total (in mol/L)
10–6 Fe2+
Rickard (2006)
0
10–4 Fe2+
Rickard (2006)
–2
–4
–6
–8
–10
4
5
6
7
8
9
10
pH
Figure 5-1. Comparison of the pH dependence of the measured total sulphide concentration from Rickard
(2006) and predicted activities of H2S or HS– for the dissolution of mackinawite for dissolved Fe(II) concentrations of 10–4, 10–6, and 10–8 mol·dm–3. The predicted concentrations are calculated from Reaction (5-1)
for pH < 7 and Reaction (5-4) for pH ≥ 7.
Table 5-1. Summary of free energy of formation data used in solubility calculations.
Species
DGf0
(kJ/mol)
Mackinawite FeSm
–98.38
Reference
Inferred from Rickard 2006
Greigite Fe3S4g
–308.30
Pyrite FeS2p
–160.229
Robie et al. 1978
Fe2+
–90.53
H2S(aq)
–27.83
Wagman et al. 1982
HS–
12.05
HS2–
22.07
Table 5-2. Free energy of reaction used in solubility calculations.
Reaction
DGr0
(kJ/mol)
FeSm + 2H+ = Fe2+ + H2S
–19.98
FeSm + H+ = Fe2+ + HS–
19.90
Fe3S4g + 6H+ = 3Fe2+ + 3H2S + S0
–46.78
Fe3S4g + 3H+ = 3Fe2+ + 3HS– + S0
72.86
FeS2p + 2H+ = Fe2+ + H2S + S0
41.87
FeS2p + H+ = Fe2+ + HS– + S0
81.75
FeS2p + H+ = Fe2+ + HS2–
91.77
26
SKB TR-13-19
4
2
log10 [S(-II)] (in mol/L)
0
–2
–4
–6
–8
–10
Experimental
Predicted
–12
–14
4
5
6
7
8
9
10
pH
Figure 5-2. Comparison of the experimental and predicted activities of H2S (for pH < 7) or HS– (for pH ≥ 7)
for the solubility of greigite for a dissolved Fe(II) concentration of 10–4 mol·dm–3. The experimental data
point is taken from Rickard and Luther (2007), calculated from the original study of Berner (1967). The
predicted concentrations are calculated from Reaction (5-5a) for pH < 7 and Reaction (5-5b) for pH ≥ 7.
Rickard and Luther (2007) report only a single experimental measurement of the solubility of greigite,
that by Berner (1967). Figure 5‑2 shows a similar comparison of measured and thermodynamically
predicted sulphide concentrations as a function of pH for the dissolution of greigite. In this case,
Fe3S4g is assumed to dissolve with the formation of sulphide and elemental sulphur S0
Fe3S4g + 6H+ = 3Fe2+ + 3H2S + S0(5-5a)
and
Fe3S4g + 3H+ = 3Fe2+ + 3HS– + S0(5-5b)
As for the solubility of mackinawite, there is reasonable agreement between the thermodynamically
predicted sulphide concentration for the dissolution of greigite and that derived from the experimental
study of Berner (1967), albeit only for a single experimental data point for comparison.
5.2Pyrite
Having established the principle of using thermodynamics to predict the solubility of mackinawite
and greigite, we now extend the method to the prediction of the solubility of pyrite for which there
are no experimental measurements. For this purpose, we consider both the dissolution of FeS2p to
produce sulphide and elemental sulphur
FeS2p + 2H+ = Fe2+ + H2S + S0(5-6a)
and
FeS2p + H+ = Fe2+ + HS– + S0(5-6b)
and the dissolution of pyrite to produce ferrous and polysulphide ions
FeS2p + H+ = Fe2+ + HS2–(5-7)
SKB TR-13-1927
Figure 5‑3 shows the predicted concentration of sulphide in equilibrium with pyrite in the presence
of elemental sulphur for dissolved Fe2+ concentrations of 10–4, 10–6, and 10–8 mol·dm–3 based on
Reactions (5-6a) and (5-6b) for pH < 7 and pH ≥ 7, respectively.
Predicting the concentration of sulphide as a function of Fe2+ concentration is appropriate if the solu
bility of some other Fe(II) phase controls the dissolved Fe(II) concentration. However, since pyrite is
likely to be the least-soluble Fe(II) solid present, it is reasonable to assume that dissolution of FeS2p
will control the dissolved [Fe2+], in which case we need to consider the congruent dissolution of pyrite.
Figure 5‑4 shows the predicted sulphide or polysulphide concentrations as a function of pH for the
congruent dissolution of pyrite for Reactions (5-6a), (5-6b), and (5-7). Because the pyrite is dissolv
ing congruently, the dissolved Fe2+ concentration also varies with pH and is equal to the concentra
tion of H2S, HS–, or HS2–. Thus, depending upon the precise dissolution pathway, the concentration
of sulphide or polysulphide in bentonite pore water (assumed to be pH 8) is predicted to be of the
order of 10–11.2 to 10–12.0 mol·dm–3 at 25°C.
As noted above, the solubility of FeS2p is too low to measure at ambient temperature. Ohmoto et al.
(1994) measured the solubility of pyrite in NaCl solutions at temperatures of 250–350°C. Their data
seem to indicate an increase in solubility by as much as two orders of magnitude over this range of
temperature, although there is a lot of scatter in the data and the authors themselves concluded that
the solubility of FeS2p and of other iron sulphides is relatively insensitive to temperature.
The solubility of pyrite is significantly lower than that for mackinawite or greigite. Figure 5‑5 shows
a comparison of the predicted HS– activity based on Reactions (5-1) and (5-4) for mackinawite,
Reactions (5-5a) and (5-5b) for greigite, and Reactions (5-6a) and (5-6b) for pyrite for a dissolved
Fe2+ concentration of 10–6 mol·dm–3. Based on this comparison, the concentration of HS– is a factor of
approximately 10 orders of magnitude lower for pyrite than for either of the other two iron sulphides.
–6
–8
log10 acvity (H2S/HS-)
–10
–12
–14
–16
–18
10–8 Fe2+
–20
10–6 Fe2+
–22
10–4 Fe2+
–24
4
6
pH
8
10
Figure 5-3. Predicted activities of H2S or HS– for the solubility of pyrite with the formation of elemental
sulphur for dissolved Fe(II) concentrations of 10–4, 10–6, and 10–8 mol·dm–3.
28
SKB TR-13-19
–6
FeS2 + 2H+ = Fe2+ + H2S + S0
log10 acvity (H2S/HS-/HS2-)
–7
FeS2 + H+ = Fe2+ + HS– + S0
FeS2 + H+ = Fe2+ + HS2–
–8
–9
–10
–11
–12
–13
–14
4
6
8
10
pH
Figure 5-4. Predicted activities of H2S, HS–, or HS2– for the congruent dissolution of pyrite.
3
log10 acvity sulphide
0
–3
–6
–9
–12
–15
Mackinawite
Greigite
Pyrite
–18
–21
4
6
pH
8
10
Figure 5-5. Comparison of the predicted sulphide activity for the dissolution of mackinawite, greigite, and
pyrite based on Reactions (5-1) and (5-4), (5-5a) and (5-5b), and (5-6a) and (5-6b), respectively. Dissolved
Fe2+ concentration of 10–6 mol·dm–3. For each iron sulphide, the first of the two noted reactions was used
for pH values < 7 and the second noted reaction for pH ≥ 7. At pH < 7, the dissolved sulphide is present
as H2S, with HS– predominant at pH ≥ 7.
SKB TR-13-1929
6
Pyrite dissolution and electrochemistry
The dissolution of pyrite can occur oxidatively, reductively, or congruently with no change in oxi
dation state. These different mechanisms are distinguished by the change, if any, in the oxidation
state of the sulphur species. Oxidative dissolution is characterised by the oxidation of sulphur from
the average –I oxidation state in FeS2p to SO42– (a net transfer of seven electrons from S22–), although
other stable intermediate oxidation states are also formed under some circumstances. Reductive
dissolution is defined here as involving the reduction of S22– to two S2– species (a net transfer of one
electron). Finally, congruent dissolution is defined as the dissolution of FeS2p as Fe2+ and S22–, with
possible subsequent disproportionation of the dissolved polysulphide species.
Of these three dissolution pathways, those of most interest here are the reductive and congruent
dissolution routes. Although oxidative dissolution is not the main focus of the current review, the
mechanism is briefly summarised here because of the impact of the oxic phase on the subsequent
behaviour during the anaerobic period.
6.1
The oxidative dissolution of pyrite is involved in a number of important industrial and environmental
processes, including: mineral flotation and leaching, desulphurisation of coal, and acid rock (or mine)
drainage (Chandra and Gerson 2010, Langmuir 1997, Rickard and Luther 2007, Rimstidt and Vaughan
2003, Vaughan 2005). Although the oxidative dissolution process can be accelerated by microbial
activity, the predominant underlying processes are electrochemical in nature, with the mechanism and
rate of dissolution primarily determined by potential and the (semi-conducting) properties of the pyrite,
as well as the pH, temperature, the nature and concentration of the oxidant, hydrodynamic conditions,
grain size, the surface area:volume ratio, and pressure (Chandra and Gerson 2010).
The two most important oxidants for pyrite are O2 and Fe3+. The overall stoichiometry of the respective
dissolution processes are (Vaughan 2005)
FeS2p + 3.5O2 + H2O → Fe2+ + 2SO42– + 2H+(6-1a)
and
FeS2p + 14Fe3+ + 8H2O → 15Fe2+ + 2SO42– + 16H+(6-1b)
Although ferric species are important intermediates of the oxidation process and are occasionally
described as products of the overall reaction, the general consensus is that oxidative dissolution
under acidic conditions produces Fe2+ species. The species undergoing oxidation, therefore, is the
disulphide S22–, with sulphate the predominant product, although some researchers report various
stable intermediate sulphur oxyanions, such as thiosulphate S2O32–, polythionates SnO62–, and sulphite
SO32– (Chandra and Gerson 2010, Moses et al. 1987). Precipitated Fe(III) corrosion products are
reported in neutral and alkaline solutions (Caldeira et al. 2003, 2010, Huminicki and Rimstidt 2009,
Todd et al. 2003), in which dissolved Fe(II) is more readily oxidised homogeneously to Fe(III) by O2.
The multiple electron transfer steps involved in the oxidation of S22– occur sequentially at the anodic
sites on the pyrite surface. Based on the results of experiments using isotopically labelled H2O and
O2, it has been shown that the oxygen atoms in the eventual SO42– (or sulphur oxyanion) product
derive from H2O, whereas the O2 participates in the cathodic process and can be found in precipi
tated iron oxyhydroxides (Heidel and Tichomirowa 2010, Usher et al. 2004). This is evidence for
anodic and cathodic processes occurring on physically separated sites.
The sequence of electron transfers involved in the oxidation of the S sites (and the addition of
oxygen from H2O) can be represented by (after Rimstidt and Vaughan 2003)
SKB TR-13-1931
+ H2O
py − S − S → py − S − S+ + e − 
→ py − S − SOH + H + 
→ py − S − SO + H + + e −
↓
+
−
+
+ H2O
py − S − SO 2 + H + e ←
 py − S − SO 2 H + H ← py − S − SO + + e −
↓
+ H2O
py − S − SO 2+ + e − 
→ py − S − S − O3 H + H + 
→ py − S − SO3 + H + + e − (6-2)
where py-S-S represents a polysulphide species at the pyrite (py) surface. This sequence represents
a series of electron transfer, hydrolysis, and deprotonation steps.
The surface intermediate py-S-SO3 is a key species in the sequential oxidation process. In alkaline
conditions or in anoxic Fe3+ solutions (Moses et al. 1987), the py-S bond can cleave and thiosulphate
S2O32– is observed in solution. In the presence of either O2 or Fe3+, especially in acidic solution, the
surface intermediate is further oxidised all the way to sulphate, which, following cleavage of the S-S
bond, is the only sulphur oxyanion observed in solution
+ H2O
py − S − SO3 → py − S − SO3+ + e − 
→ py − S − SO 4 H + H + 
→ py − S − SO 4 + H + + e −
↓
py + S2 O
↓
2−
3
py − S + SO 24 −
(6-3)
Reactions (6-2) and (6-3) represent one possible pathway for the oxidation of the polysulphide. The
key aspects are that: (i) oxidation occurs sequentially involving the addition of O from H2O and the
removal of electrons, (ii) the reaction occurs on anodic sites, independent of the cathodic process or
the involvement of the oxidant, and (iii) a range of possible surface and dissolved intermediate spe
cies are possible (Borda et al. 2003, 2004).
Druschel and Borda (2006) present a schematic of a range of possible pathways for the oxidative
dissolution of pyrite (Figure 6‑1). Pathways 1A and 1B in the figure are similar to those described
above in which the Fe-S bond breaks to produce S2O32– (which can then react further to produce SO42–,
Pathway 1A) or in which the S-S bond breaks to ultimately form sulphate, possibly through a sulphite
SO32– solution intermediate (Pathway 1B). Pathway 2 is interesting in the current discussion because
it involves congruent dissolution of the polysulphide, which is then postulated to be oxidised in solu
tion. Finally, Pathway 3 involves photochemical processes.
Figure 6-1. Schematic of different pathways for the oxidative dissolution of pyrite (Druschel and Borda
2006). See text for explanation of different pathways.
32
SKB TR-13-19
The cathodic reaction involves the overall reduction of Fe(III) or O2 (or of other oxidants, discussed
below). In the case of the oxidative dissolution of pyrite in O2– or Fe3+-containing solution, the actual
oxidant is believed to be Fe3+, either present in solution or formed by the homogeneous oxidation of
Fe(II) by O2. The cathodic reaction is believed to occur at Fe(II) sites on the pyrite surface (Rimstidt
and Vaughan 2003). Electron transfer may be facilitated by shuttling of the Fe(II) site between Fe(II)
and Fe(III) states. Although the cathodic reaction involves the transfer of fewer electrons and involves
fewer steps, it is believed to be rate controlling.
A common observation is that the dissolved [STOT]:[FeTOT] is < 2, even in acid solution in which the
extent of precipitation of alteration products would be expected to be minimal (Descostes et al. 2004,
2010). This lead Descostes et al. (2004) to propose an alternative mechanism that includes the pre
cipitation of elemental S
FeS2p + 2.9O2 + 0.6H2O → Fe2+ + 0.4S0(s) + 1.6SO42– + 1.2H+(6-4)
where the stoichiometry was based on the observed [STOT]:[FeTOT] of 1.6. Alternatively, Fe2+ sorption
on the pyrite surface is also possible (Descostes et al. 2010).
Williamson and Rimstidt (1994) reviewed data from various kinetic studies of the rate of oxidative
dissolution of pyrite by O2 and by Fe3+ (under oxic and anoxic conditions). The best-fit rate law for
O2 is
Rate(mol m −2 s −1 ) = 10−8.19( ±0.10)
m O0.5(2 ±0.04)
±0.01)
m 0.11(
H+
(6-5)
where m is the molality in mol kg–1. This expression is valid over the pH range 2–10 and for dis
solved oxygen concentrations of 7·10–7 to 2·10–2 mol·dm–3 (0.02 to 620 mg/L). The corresponding
rate expressions for oxidation by Fe3+ are
Rate (mol m −2 s −1 ) = 10−8.58( ±0.15)
±0.02)
m 0.30(
Fe3+
±0.03) 0.32( ±0.04)
m 0.47(
m H+
Fe2+
(6-6)
under anoxic conditions and
±0.07)
m 0.93(
Fe3+
Rate (mol m −2 s −1 ) = 10−6.07( ±0.57) 0.40(
(6-7)
m Fe2+ ±0.06)
in the presence of O2. The fractional reaction order with respect to the dissolved O2 concentration
in Equation (6-5) is consistent with an electrochemical mechanism.
A detailed discussion of the mechanism of the oxidative dissolution of pyrite is outside of the scope
of the current review. In addition to the extensive literature on the dissolution behaviour in abiotic
acidic environments, the effect of a number of other parameters have been considered:
• Oxidants other than O2 or Fe3+, including UV (Borda et al. 2003, 2004) and visible (Schoonen
et al. 2000) light, H2O2/OH radicals (Schoonen et al. 2010), γ‑radiation (Lefticariu et al. 2010),
MnO2 (Schippers and Jørgensen 2001, 2002), NO3– (Schippers and Jørgensen 2002), and amor
phous precipitated Fe(III) (Schippers and Jørgensen 2002).
• Neutral and/or alkaline pH (Caldeira et al. 2003, 2010, Moses et al. 1987, Moses and Herman
1991, Todd et al. 2003, Williamson and Rimstidt 1994).
• Microbial effects (Bosch and Meckenstock 2012, Bosch et al. 2012, Brunner et al. 2008, Gleisner
et al. 2006, Jørgensen et al. 2009, Lizama and Suzuki 1989, Rawlings et al. 1999, Schippers and
Jørgensen 2001, 2002, Taylor et al. 1984, Torrentó et al. 2010).
• Temperature (Schoonen et al. 2000).
• Carbonate/bicarbonate (Caldeira et al. 2003, 2010, Ciminelli and Osseo-Asare 1995, Descostes
et al. 2002, Huminicki and Rimstidt 2009, Nicholson et al. 1988, 1990).
• Ionic strength and Cl– and SO42– concentration (Williamson and Rimstidt 1994).
• Unsaturated or atmospheric environments (Jerz and Rimstidt 2004, Todd et al. 2003).
SKB TR-13-1933
For the current report, it is important to note that the oxidative dissolution of pyrite can occur under
anoxic conditions (sometimes, confusingly, referred to by some authors as dissolution under “anaerobic”
conditions). The oxidative dissolution of FeS2p by Fe3+ does occur under anoxic conditions, but is
typically only sustained if O2 is present to homogeneously oxidise Fe2+ to Fe3+. Ultraviolet and vis
ible light increases the rate of oxidative dissolution in the presence of Fe3+ or O2 (Borda et al. 2003,
2004, Schoonen et al. 2000), although there is no evidence that either would sustain dissolution in
the absence of the other oxidants in the system. On the other hand, γ‑irradiation alone will cause
oxidative dissolution by the reaction of OH radicals, although Fe3+ species are inevitably produced
in solution because of the irradiation (Lefticariu et al. 2010). Schippers and Jørgensen (2001, 2002)
studied the effect of MnO2, NO3–, and precipitated amorphous Fe(III) oxide on the oxidative dis
solution of pyrite in anoxic marine sediments containing natural microbial populations. Manganese
dioxide was found to oxidise FeS2p under both biotic and abiotic conditions. However, neither NO3–
nor amorphous Fe(III) oxide acted as oxidants for pyrite under biotic or abiotic conditions, although
both species did oxidise iron sulphide (FeS) in the presence of microbes, clearly showing the greater
stability of pyrite compared with FeS.
The pH of pore water in the buffer and backfill materials in the repository is expected to be slightly
alkaline because of buffering by calcite mineral impurities. As such, the dissolution behaviour of
pyrite at neutral or moderately alkaline pH is of more interest than that at low pH designed to simu
late acid rock drainage conditions. Despite its lower solubility, there is evidence that Fe3+ still acts
an the oxidant in neutral and alkaline solution, although O2 is required to sustain the reaction (Moses
and Herman 1991, Moses et al. 1987). Precipitation of alteration products on the pyrite surface is
also an issue in neutral and alkaline solution (Caldeira et al. 2003, 2010, Huminicki and Rimstidt 2009,
Todd et al. 2003). The alteration products tend to be ferric species, including precipitated Fe2(SO4)3
and FeOH(SO4), with FeOOH or Fe(OH)3 reported with increasing pH. These species tend to result
in a decrease in dissolution rate with time because the precipitated film restricts O2 transport to the
pyrite surface (Huminicki and Rimstidt 2009).
Microbes are known to accelerate the dissolution of pyrite in acidic solutions by a factor of up to 106
(Vaughan 2005). Species such as Thiobacillus ferrooxidans facilitate the oxidation of both Fe2+ and
sulphur. Gleisner et al. (2006) report an indirect effect of microbial activity through the oxidation of
Fe2+ to Fe3+, but report no enhancement in the rate. As noted above, Schippers and Jørgensen (2001,
2002) found that naturally occurring microbes in anoxic sea sediments did not catalyse the oxidation
of pyrite by NO3– or amorphous Fe(III) oxide. However, recently the oxidation of pyrite supported
by the microbial reduction of nitrate has been reported (Bosch and Meckenstock 2012, Bosch et al.
2012, Jørgensen et al. 2009, Torrentó et al. 2010).
Jerz and Rimstidt (2004) studied the oxidation of pyrite in humid air. At a constant relative humidity
(RH) of 96.7%, the dissolution rate was found to decrease with time, which was attributed to increasing
transport control by O2 diffusion across a thickening aqueous film containing FeSO4 and H2SO4. Because
the RH was controlled, the solution layer thickened as dissolution proceeded as the solution com
position is constant at a given RH. At RH values less than 95%, a precipitated ferrous sulphate film
formed, presumably because this value is below the deliquescence RH for iron sulphate. Preferential
precipitation at grain boundaries caused the pyrite crystals to disaggregate.
6.2
The reductive dissolution of pyrite is of interest because it would not only release sulphide but also
result in the formation of a form of iron sulphide more soluble than pyrite, such as pyrrhotite or troilite.
Reduction of pyrite is known to occur at elevated temperatures in H2 atmospheres (Lambert et al.
1980, 1998). In inert atmospheres, pyrite is reduced at high temperatures to pyrrhotite along with
the formation of elemental sulphur. In the presence of H2, both pyrrhotite and H2S are formed, the
overall reaction being (Hol et al. 2010, Lambert et al. 1998)
FeS2p + (1–x)H2 → FeS1+x + (1–x)H2S(6-8)
However, the kinetics of the reaction are not significant at temperatures below 300–400°C and the
process exhibits a strong temperature dependence with an activation energy of 90 kJ/mol (Lambert
34
SKB TR-13-19
et al. 1998). Hol et al. (2010) tried to induce the reduction of pyrite at more moderate temperatures
(35°C and 55°C) and H2 pressures (0.11–0.14 MPa) using bacterial communities, including sulphate
reducers. Despite suppressing unwanted microbial reduction processes, such as methane formation
and sulphate reduction, no reduction of pyrite was observed.
As part of the study of the impact of repository conditions on the pyrite contained in Callovo-Oxfordian
clay in the French nuclear waste management programme, Truche et al. (2010) have demonstrated
the reductive dissolution of pyrite with H2 at temperatures between 90 and 180°C. Experiments were
performed in a dilute NaCl solution (0.027 mol·dm–3) at pH 6.9–8.7 and buffered by crushed calcite.
Reduction of FeS2p to FeS1+x (0 < x < 0.125) was demonstrated for H2 partial pressures of 0.8 MPa and
1.8 MPa. Figure 6‑2 shows the results of selected experiments at temperatures of 90°C and 150°C,
for H2 partial pressures of 0 MPa and 0.8 MPa, and for different sizes of pyrite particles (with corre
spondingly different reactivity). Dissolved HS– is observed at both 90°C and 150°C with 0.8 MPa H2
for the finer (and more reactive) pyrite particle size. Although no HS– was observed for the coarsergrained material at 90°C, sulphide was observed in other experiments with this material at higher
temperature and/or H2 pressure. The [HS–] appears to reach a constant value, possibly reflecting an
equilibrium solubility. The increase in dissolution rate with temperature is equivalent to an activation
energy of 53 kJ/mol. No HS– is observed at 150°C in the absence of H2, even with the finer-grained
material.
Truche et al. (2010) developed an expression for the release of HS– (N in mol m–2) as a function of
time (t in hours), H2 pressure (pH2 in Pa) and temperature (T in K)
log N = −5.22 + 0.47 log t + 1.10log p H2 −
2755
(6-9)
T
The pyrrhotite formed by the reduction process forms a crust over the pyrite core, with the rate jointly
controlled by the pyrite reduction process and the diffusion of HS– through the surface film. Betelu
et al. (2012) have shown that the rate of reductive dissolution in the presence of H2 can be increased
by cathodic polarisation.
Hydrogen is not the only reductant that may lead to the reductive dissolution of pyrite. Luther (1987)
has shown that the experimentally reported reduction of FeS2p by Cr(II) is consistent with molecular
orbital theory considerations.
3.0
Dissolved HS– concentraon (mmol/L)
90°C, 0.8 MPa H2, 40–80 µm FeS2
90°C, 0.8 MPa H2, <40 µm FeS2
2.5
150°C, 0.8 MPa H2, <40 µm FeS2
150°C, 0 MPa H2, <40 µm FeS2
2.0
1.5
1.0
0.5
0
0
100
200
300
400
500
600
Time (hours)
Figure 6-2. Time dependence of the dissolved sulphide concentration as the result of the reductive dissolution
of pyrite by hydrogen. Based on data by Truche et al. (2010).
SKB TR-13-1935
6.3
Very few studies of the chemical dissolution of FeS2p have been published. Although Druschel and
Borda (2006) speculate on a polysulphide oxidative dissolution pathway that essentially involves
a chemical dissolution process followed by homogeneous oxidation of the dissolved polysulphide
(Pathway 2, Figure 6‑1), these authors provide no direct evidence in support of such a mechanism.
The indirect evidence suggested for such a pathway includes the observation of elemental sulphur
reported by a number of authors and solution [SO42–]:[Fe] ratios less than two, the latter evidence
implying that not all of the dissolved polysulphide is subsequently fully oxidised to sulphate.
Weerasooriya and Tobschall (2005) report dissolution of pyrite in anaerobic solutions. Dissolved iron
was measured in acidified 0.1 mol·dm–3 NaClO4 solutions at pH < 4.5, which was ascribed to a surface
complexation process that resulted in the release of Fe2+. The authors report both H2S in acidic solution
and S22– in alkaline solution (>pH 10), although there is no description of how the latter was measured.
The detection of sulphide in solution was taken as an indication of pyrite dissolution. However, the
exact experimental details are not clear, since the authors refer to the effect of O2 in the text and it
is not clear whether the data shown were measured under anaerobic conditions as implied in the
Abstract. No dissolution rates were given and equilibration times were limited to 30 minutes.
Thomas et al. (2001) report pyrite dissolution rates in argon-purged 0.1 mol·dm–3 perchloric acid (pH 1)
solution. A peak Fe release (dissolution) rate of 6·10–9 mol m–2 s–1 (0.2 mol m–2 y–1) was reported after
4 h exposure, which then decreased to zero by the end of the 28-h-duration experiment.
6.4
Electrochemistry of pyrite
There is an extensive literature on electrochemical measurements on pyrite. However, the majority of
studies have focussed on the oxidation of pyrite, primarily in acidic solution, and are not discussed in
detail here. Among the studies that have been performed are the following:
• Mechanism and electrochemical kinetics (Biegler and Swift 1979, Hamilton and Woods 1981,
Holmes and Crundwell 2000, Kelsall et al. 1999, Liu et al. 2008, 2009, Mycroft et al. 1990).
• Semi-conductivity, impurity content, and reactivity (Biegler 1976, Cruz et al. 2001, Kelsall et al.
1999, Lehner et al. 2007).
• Film growth and characterisation (Giannetti et al. 2006, Mycroft et al. 1990).
• The oxygen reduction reaction (Ahlberg and Broo 1996, Biegler 1976, Rand 1977).
• Measurement of the open-circuit potential (Moslemi et al. 2011).
• The effect of microbes on pyrite oxidation (Cabral and Ignatiadis 2001, Holmes et al. 1999).
The electrochemical reduction of pyrite has been proposed as a method for desulphurising coal
(Zhao et al. 2005, 2008). Tao et al. (1994, 2003) have studied the electrochemical behaviour of freshly
fractured pyrite in deaerated borate buffer solution at pH 9.2. Figure 6‑3 shows a cyclic voltammogram
of pyrite from a Chinese coal sample measured in deaerated pH 9.2 borate buffer solution at room
temperature. The authors identified three anodic peaks (labelled I, II, and III) and three cathodic peaks
(labelled IV, V, and VI) on the voltammogram. Tao et al. (2003) assigned these peaks to the following
processes:
I
Fe + 2OH– → Fe(OH)2 + 2e–(6-10)
II
initial oxidation of FeS2p(6-11)
IIIFeS2p + 11H2O → Fe(OH)3 + 2SO42– + 19H+ + 15e–(6-12)
IV(n–1)Fe(OH)3 + FeSn + 3(n–1)e– → nFeS + 3(n–1)OH–(6-13)
VS0 + H+ + 2e– → HS–(6-14)
VIFeS2p + H2O + 2e– → FeS + HS– + OH–(6-15a)
FeS2p + 2H2O + 4e– → Fe + 2HS– + 2OH–(6-15b)
36
SKB TR-13-19
Figure 6-3. Cyclic voltammetric behaviour of a pyrite rotating disc electrode in borate buffer at pH 9.2
and the corresponding current for a gold ring at a potential of +0.25 VSHE (Tao et al. 2003). Although the
authors suggest both 1st and 5th cycles are shown, the curves for the latter are not visible on the original.
Electrode rotation rate 2,000 rpm, potential scan rate not defined.
where the Fe oxidised in Reaction (6-10) and the sulphur reduced in Reaction (6-14) were formed
during the preceding cathodic and anodic scans, respectively. Cathodic feature VI is accompanied
by a large ring oxidation current, which Tao et al. (2003) attributed to the oxidation of HS– to S0.
The assignment of these peaks is based on earlier studies by Ahlberg et al. (1990) and Hamilton and
Woods (1981). However, little supporting evidence is provided in these studies for the assignment of
the different peaks and, indeed, Hamilton and Woods (1981) reported fewer voltammetric features
and Ahlberg et al. (1990) report a greater number of anodic and cathodic peaks than observed by
Tao et al. (2003). Therefore, the assignment of the different peaks is somewhat uncertain and, in
particular, that for the reduction of FeS2p (peak VI in Figure 6‑3), which Ahlberg et al. (1990) report
as beginning at approximately –0.8 VSCE at pH 11. Ahlberg et al. (1990) did not have the advantage
of using a rotating ring-disc electrode, but one must wonder whether the corresponding ring oxida
tion current observed by Tao et al. (2003) may not be the result of the oxidation of H2 produced by
the reduction of H2O on the disc electrode at potentials more-negative than –0.65 VSHE (the reversible
potential for the evolution of H2 is approximately –0.54 VSHE at pH 9.2). Nevertheless, the potential
of –0.65 VSHE for the onset of measurable pyrite reduction is consistent with the equilibrium potential
for Reaction (6-15a). Using the free energy of formation value for FeS2p and HS– in Table 5‑1 and
corresponding values for pyrrhotite and troilite of –98.9 kJ/mol and –102.9 kJ/mol, respectively
(Robie and Hemingway 1995), the standard potential E0 for Reaction (6-15a) is –0.794 VSHE and
–0.774 VSHE for pyrite reduction to pyrrhotite and troilite, respectively. At pH 9.2, the equilibrium
potentials for this reaction are –0.475 VSHE and –0.455 VSHE for pyrrhotite and troilite, respectively.
Thus, in electrochemical terms, compared with the potential of –0.65 VSHE at which Tao et al. (2003)
report a measurable pyrite reduction current, the process exhibits an overpotential 175–200 mV.
Tao et al. (1994, 2003) report the “stable” potential of freshly fractured pyrite to be –0.28 VSHE in
deaerated borate buffer at pH 9.2. This potential was the value at which freshly fractured pyrite
surfaces exhibited no net current (i.e., neither oxidation nor reduction) and which, therefore, is a
measure of what is more commonly referred to as the open-circuit or corrosion potential (ECORR). If
we take this value as representative of ECORR of pyrite under anaerobic conditions, then the data in
Figure 6‑3 suggests that we need to polarise the surface by approximately –0.4 V in order to cathodi
cally reduce FeS2p to FeS via Reaction (6-15a), if we assume that peak VI corresponds to the reduction
of pyrite. Thus, in order to reductively dissolve pyrite under freely corroding conditions (as might
occur in the repository), the presence of a reducing species with a sufficiently negative equilibrium
potential to cathodically polarise the potential of the pyrite by 0.4 V to –0.65 VSHE is required.
SKB TR-13-1937
By way of example, let us assume that H2 is the reductant and that the reduction of pyrite is accom
panied by the oxidation of hydrogen
H2 → 2H+ + 2e–(6-16)
which is essentially the process occurring in the work of Truche et al. (2010), since it is believed to
follow an electrochemical mechanism (Betelu et al. 2012). The H2 pressure required to produce an
equilibrium potential of –0.65 VSHE at pH 9.2 and 25°C is estimated to be 350 MPa. Truche et al. (2010)
did not perform experiments at temperatures below 90°C, but the estimated H2 pressure is not inconsist
ent with their findings.
38
SKB TR-13-19
7
Implications for repository performance
Various aspects of the properties of pyrite are important for the performance of the repository
system, including:
• The initial state of the pyrite in the bentonite.
• The evolution of the pyrite behaviour in the buffer and backfill, especially the release of sulphide
species and the extent of oxidative, reductive, and chemical dissolution.
• The consequences for corrosion of the canister, especially the (beneficial) consumption of O2 and
the (detrimental) formation of dissolved sulphur species that might support corrosion during the
anoxic period.
7.1
Initial state of pyrite in bentonite
There is no information on the initial state of the pyrite impurities in the bentonite clay, other than
the likely range of the pyrite content. If we assume that the pyrite was formed by a low-temperature
dissolution-precipitation mechanism, as in the case of the pyrite inclusions in the bentonite sample
from Zao, Japan (Fukushi et al. 2010), then the pyrite is more likely to exhibit p-type semi-conduct
ing properties than n-type, based on the evidence summarised in Section 2.1.3.
During exposure to the atmosphere during mining and milling operations, the pyrite could undergo
oxidative dissolution if moisture is present, although the existence of pyrite in the processed clay
suggests that the extent of such processes are limited. Since, the system is highly pH-buffered (because
of the presence of calcite), the pyrite surface is likely to become encrusted with alteration products
(Section 6.1). If oxidation occurs under unsaturated (atmospheric) conditions, the precipitated phase
may be FeSO4 (Jerz and Rimstidt 2004). If oxidation occurs under saturated conditions, then the
alteration products are more likely to include ferric species, such as Fe2(SO4)3 and FeOH(SO4), with
FeOOH or Fe(OH)3 possible with increasing pH (Caldeira et al. 2003, 2010, Huminicki and Rimstidt
2009, Todd et al. 2003).
However, because of the lack of direct characterisation of the pyrite particles in bentonite clays, this
suggested initial state of the pyrite should be treated cautiously.
7.2
Evolution of pyrite behaviour in repository
7.2.1 In the buffer
Pyrite in the buffer material will be exposed first to warm, oxic conditions and then, as the repository
environment evolves, an indefinite cool, anoxic period. Because microbial activity is suppressed
virtually completely in highly compacted bentonite, pyrite will only react abiotically in the buffer.
During the oxic phase, the pyrite will undergo similar processes to those described above during
the mining and milling operations. During the post-closure period, however, the temperature will
be higher and, depending upon the permeability of the host rock, the buffer may be saturated. These
reactions will consume a fraction of the initially trapped atmospheric O2, resulting in the formation
of sulphate and the dissolution of calcite in response to the generation of H+. The pyrite particles will
become further encrusted with a layer of alteration products, most likely ferric species. The presence
and further growth of this outer layer will reduce the rate of oxidation over that observed under acidic,
film-free conditions.
In terms of the release of sulphur species during the oxic period, the predominant species will be
sulphate. Depending upon the pathway for pyrite oxidation (Figure 6‑1), small amounts of other
sulphur oxyanions may be released into solution, such as thiosulphate S2O32–, polythionates SnO62–,
and sulphite SO32– (Chandra and Gerson 2010, Moses et al. 1987). However, if O2 is still present in
the buffer pore water, then these species may be homogeneously oxidised further.
SKB TR-13-1939
Once anoxic conditions become established, the pyrite is likely to react very slowly, if at all. In the
absence of microbial activity or a high pressure of H2, the pyrite will be neither further oxidised nor
undergo reductive dissolution.
Truche et al. (2010) have shown that pyrite can be reduced to pyrrhotite by H2 at elevated temperatures,
accompanied by the release of sulphide. However, there is no evidence that this reaction occurs at
any significant rate at the lower temperatures expected during the anoxic period. Furthermore, there
will be limited amounts of H2 produced in the buffer material. Corrosion of copper by HS–, with
the accompanying formation of H2, will only result from the diffusion of HS– from the backfill or
groundwater. The rate of this process is estimated to be equivalent to a corrosion rate of < 1 nm/y
(King et al. 2011a, b), and at this rate the H2 produced will diffuse away from the canister as dis
solved H2 and will not accumulate as a H2 gas phase of elevated pressure (King 2012).
Despite the fact that reductive dissolution of FeS2p by H2 is unlikely to occur in the buffer, it is inter
esting to apply the sulphide production rate expression (Equation (6-9)) developed by Truche et al.
(2010) to the “what-if” scenario considered by SKB (2010a). In this scenario, it was assumed that
the copper is corroded by water as proposed by Szakálos et al. (2007) and that an equilibrium H2
partial pressure of 0.001 bar is attained. Let us further assume that Equation (6-9) is valid at this H2
partial pressure and at a temperature of 80°C (the maximum canister temperature), although these
conditions fall outside those investigated by Truche et al. (2010). We can then use Equation (6-9) to
predict the quantity of HS– produced by the reductive dissolution of pyrite by H2 as a function of time.
Figure 7-1 shows the predicted amount of HS– produced by the reductive dissolution of pyrite
expressed in terms of moles per m2 pyrite surface and per borehole.1 The rate of sulphide production
is low and even after 1 million years amounts to only 0.008 mol per borehole. For the reaction
2Cu + HS– + H2O → Cu2S + H2 + OH–(7-1)
the predicted depth of corrosion due to this source of sulphide is 6 nm.
Quanty of HS- (mol/m2 or mol/borehole)
10–2
Amount per borehole
Amount per m2 pyrite
10–3
10–4
10–5
10–6
10–7
10–8
10–9
1
10
100
1,000
10,000
100,000
1,000,000
Time (years)
Figure 7-1. Estimated production of sulphide by the reductive dissolution of pyrite by hydrogen for a H2
pressure of 0.001 bar and a temperature of 80°C (based on data of Truche et al. 2010).
For this calculation, it was assumed that the pyrite exists as spherical particles of radius 10 µm and, hence,
a surface area:volume ratio of 3·104 cm–1. It was further assumed that the MX-80 bentonite contains 0.07 wt.%
FeS2p (SKB 2010a), that the canister has a length of 4.835 m and a radius of 0.525 m (SKB 2010a), the borehole
is 6.68 m deep and 1.77 m diameter (SKB 2010b), and is filled with compacted bentonite with an average dry
density of 1,600 kg m–3 (SKB 2011). The density of pyrite was taken to be 5 g cm–3 (CRC 1982). Based on these
values, the borehole is estimated to contain 23,000 kg bentonite containing 16 kg pyrite or 133 mol with a
volume of 3,200 cm3. The surface area of this pyrite is 104 m2 per borehole.
1 40
SKB TR-13-19
In the absence of oxidative or reductive dissolution, the pyrite, or more-accurately, the Fe(III)encrusted pyrite particles will be subject to chemical dissolution only. At the expected pore-water
pH of 7–8 both Fe(III) and the pyrite itself have very low solubility. In the case of pyrite, the solu
bility (expressed in terms of the HS– or HS2– concentration) is of the order of 10–11–10–12 mol·dm–3
(Figure 5‑4). The dissolution rate will be determined by the rate at which the dissolved sulphide or
polysulphide can diffuse away from the pyrite particle.
7.2.2 In the backfill
The behaviour of pyrite in the backfill will be similar to that in the buffer material, except that there
is the possibility of microbially-mediated dissolution processes.
Under oxic conditions, therefore, the oxidative dissolution of pyrite may be enhanced by the action
of Thiobacillus ferrooxdians or similar microbes (Gleisner et al. 2006, Vaughan 2005). Microbial
enhancement will increase the rate at which O2 is consumed but will not change the mechanism, since
the role of T. ferrooxidans is to promote the oxidation of Fe2+ to Fe3+ rather than to fundamentally
change the mechanism. Thus, as in the buffer, oxidative dissolution will largely be accompanied by
the formation of sulphate, with the formation of limited amounts of other sulphur oxyanions possible,
depending upon the pathway for pyrite oxidation.
During the long-term anoxic period, further oxidation of the pyrite is possible if suitable electron
acceptors and microbes are present. Nitrate reduction, promoted by species such as T. denitrificans
(Bosch et al. 2012), has been shown to support the oxidation of pyrite (Bosch and Meckenstock
2012, Bosch et al. 2012, Jørgensen et al. 2009, Torrentó et al. 2010), but there is no apparent source
of nitrate in the backfill. It is interesting to note that, despite attempts, the microbially mediated
reduction of amorphous precipitated Fe(III), as might be present in the crust of alteration products
on the partially oxidised pyrite particles, has not be shown to similarly support pyrite oxidation
(Schippers and Jørgensen 2001, 2002).
It is significant that, although microbial activity has been shown to promote the oxidation of pyrite
under anoxic conditions, there is no evidence that anaerobes can promote the reductive dissolution
of pyrite (Hol et al. 2010).
Reductive dissolution promoted by H2 is theoretically possible in the backfill during the anoxic
period but, as in the case of the buffer, there is no obvious source of H2 in the repository (other than
the anaerobic corrosion of any rock bolts) and the temperatures in the backfill will be significantly
lower than those at which Truche et al. (2010) observed pyrite reduction. Furthermore, since micro
bial activity is possible in the backfill, any H2 that does form will likely be rapidly consumed by
sulphate-reducing bacteria (King et al. 2010).
Thus, as for the buffer material, the most likely dissolution mechanism of the pyrite in the backfill
during the long-term anaerobic period is slow chemical dissolution.
7.3
Consequences for corrosion of the canister
Pyrite can influence the corrosion behaviour of the canister at all stages in the evolution of the
repository environment. During the oxic phase, pyrite will undergo oxidative dissolution resulting
in the consumption of O2, the latter then not being available to support the corrosion of the canister.
The predominant oxidation product will be sulphate SO42– which, in the absence of microbial activity,
is inert. In addition, however, small amounts of other sulphur oxyanions, such as thiosulphate S2O32–
and tetrathionate S4O62–, and polysulphides Sn2– may be produced. Macdonald and Sharifi-Asl (2011)
have reported an extensive thermodynamic analysis of possible reactions between Cu and a large
number of sulphur species. Table 7‑1 summarises a number of the reactions considered and indi
cates whether the particular sulphur species was found to “activate” copper or not, a term used by
Macdonald and Sharifi-Asl (2011) to indicate if corrosion was possible. In general, sulphide species,
polysulphides, polythionates (SxO62–) and thiosulphate were found to “activate” corrosion, whereas
polythiosulphates (SxO32–, x = 3–6) were found not to.
SKB TR-13-1941
When considering the net effect of these reactions on the corrosion of the canister, we must remem
ber that oxidant, in the form of the initially trapped atmospheric O2, is consumed in producing these
sulphur species. Thus, when carrying out a mass-balance calculation to determine how much corro
sion occurs, we must be careful not to double-count the available electron acceptors. For example,
thiosulphate can oxidise copper according to
2Cu + S2O32– = Cu2S + SO32–(7-2)
However, O2 is consumed in the formation of the S2O32– via the oxidation of pyrite
FeS2p + 1.5O2 → S2O32– + Fe2+(7-3)
where Reaction (7-3) represents the overall stoichiometry of the process but not necessarily the detailed
mechanism, which involves the interaction of H2O with polysulphide groups at anodic sites on the
pyrite surface.
The net effect of Reactions (7-2) and (7-3), however, is that 1.5 mol O2 results in the corrosion of
only 2 mol Cu, whereas the oxidation reaction
6Cu + 1.5O2 → 3Cu2O(7-4)
would result in the consumption of three times as much copper. Thus, even though some of the
intermediate sulphur oxyanions may themselves cause corrosion of the copper, there is still a net
consumption of the available oxidant as a result of the oxidative dissolution of pyrite.
In some cases, however, the sulphur species can promote corrosion with the evolution of H2 from
the reduction of H2O. For example, consider the polysulphide species S42– which reacts with copper
as follows (Table 7‑1)
8Cu + S42– + 2H+ = 4Cu2S + H2(g)(7-5)
The precise mechanism by which the polysulphide species might form is uncertain, but it involves
the oxidation of sulphur in pyrite from a mean oxidation state of –1 to a mean oxidation state of –0.5
in S42–. If we assume the oxidant is O2, then 0.5 mol O2 will produce 1 mol S42–. This polysulphide can
then react with 8 mol Cu (Reaction (7-5)), four times more than the equivalent 0.5 mol O2 would via
Reaction (7-4).
A detailed analysis of all of the reactions investigated by Macdonald and Sharifi-Asl (2011) and a
proper accounting for the electron-balance is beyond the scope of the current report. However, it
is likely that only small amounts of these “activating” sulphur species will be produced during the
oxidative dissolution of pyrite and that the major product will be the (kinetically) inert sulphate ion.
Under anaerobic conditions, all three forms of dissolution (oxidative, reductive, and chemical) are
theoretically possible, although are unlikely to occur to a significant extent. Continued oxidative dis
solution is possible in the backfill if there is a suitable source of electron acceptors (e.g., NO3–) and if
the other requirements for sustained microbial activity are met. However, there is no obvious source
of large amounts of NO3– in the backfill and any H2 that might be present will likely be consumed by
sulphate-reducing bacteria.
Reductive dissolution by H2 seems unlikely to occur in either the buffer or backfill as there is no
source of the high H2 pressures required to sustain significant pyrite reduction.
Therefore, the only viable mechanism for the release of sulphide or polysulphide during the long-term
anaerobic phase is chemical dissolution. The extent of dissolution will be limited, however, by the low
solubility of pyrite and the slow transport of dissolved HS– or HS2– from the pyrite to the canister.
In summary, the pyrite in the buffer and backfill is unlikely to be a significant source of reactants for
the copper canister. The amount of potentially aggressive dissolved oxidation products is expected
to be small and the availability of sulphide and HS2– to support corrosion under anaerobic conditions
will be limited by the extremely low solubility of FeS2p and the restrictive mass-transport conditions
in the bentonite. Thus, pyrite will only adversely impact the corrosion behaviour of the canister if the
small amounts of sulphur species promote highly localised forms of corrosion.
42
SKB TR-13-19
Table 7-1. Selection of reactions between copper and various sulphur species studied
by Macdonald and Sharifi-Asl (2011).
Reaction
Does the reactant act as an “activator”?*
+
2Cu + H2S2O3 = Cu2S + SO3 + 2H
Yes
2Cu + H2S2O4 = Cu2S + SO42– + 2H+
Yes
2Cu + HS2O3– = Cu2S + SO32– + H+
Yes
2Cu + HS2O4– = Cu2S + SO42– + H+
Yes
2–
2Cu + S2– + 2H+ = Cu2S + H2(g)
Yes
2Cu + H2S = Cu2S + H2(g)
Yes
2Cu + S2O32– = Cu2S + SO32–
Yes
2Cu + S2O42– = Cu2S + SO42–
Yes
4Cu + S22– + 2H+ = 2Cu2S + H2(g)
Yes
4Cu + S3O32– = 2Cu2S + SO32–
No
4Cu + S4O62– = 2Cu2S + SO32– + SO3(aq)
Yes
6Cu + S32– + 2H+ = 3Cu2S + H2(g)
Yes
6Cu + S4O32– = 3Cu2S + SO32–
No
6Cu + S5O62– = 3Cu2S + SO32– + SO3(aq)
Yes
8Cu + S42– + 2H+ = 4Cu2S + H2(g)
Yes
8Cu + S5O32– = 4Cu2S + SO32–
No
8Cu + S6O62– = 4Cu2S + SO32– + SO3(aq)
Yes
10Cu + S52– + 2H+ = 5Cu2S + H2(g)
Yes
10Cu + S6O32– = 5Cu2S + SO32–
No
10Cu + S7O62– = 5Cu2S + SO32– + SO3(aq)
Yes
12Cu + S62– + 2H+ = 6Cu2S + H2(g)
Yes
12Cu + S7O32– = 6Cu2S + SO32–
No
4Cu + HS3O3– = 2Cu2S + SO32– + H+
Yes
* A term used by Macdonald and Sharifi-Asl (2011) to indicate whether the sulphur species promotes corrosion of copper.
SKB TR-13-1943
8
Summary and conclusions
A review has been conducted of the properties and behaviour of pyrite and of the implications for the
corrosion of copper canisters in a KBS-3 type repository.
Pyrite (FeS2p) is an Fe(II) polysulphide with a cubic structure and exhibits semi-conducting properties,
determined primarily by the nature and extent of doping by cationic impurity elements. Pyrite is formed
by either precipitation from high-temperature melts or by aqueous-based dissolution-precipitation
processes at lower temperatures. Pyrite is the most thermodynamically stable of a number of differ
ent iron sulphides found in nature.
Bentonite clays typically contain pyrite as a minor constituent, with a content of less than approxi
mately 1 wt.%. Some sources of bentonite, however, contain little or no pyrite. Little is known about
the composition or properties of the pyrite in the bentonite clays proposed for use in the repository.
The speciation of sulphide and polysulphide in aqueous solution is well understood and the relative
distribution of species as a function of pH is known.
The solubility of pyrite is so low that it cannot be experimentally measured. Instead, the concen
tration of dissolved species in equilibrium with the solid can be estimated based on estimates of
the free energy of reaction of assumed dissolution processes. At pH 7–8, representative of that of
bentonite pore water, the concentration of HS– or HS2– in equilibrium with FeS2p is of the order of
10–12–10–11 mol·dm–3 at 25°C, approximately ten orders of magnitude lower than the corresponding
concentration of HS– in equilibrium with mackinawite FeSm or greigite Fe3S4g at the same pH.
Pyrite can dissolve oxidatively, reductively, or chemically (congruently). These mechanisms are dis
tinguished by the change in oxidation state of the sulphur, which either increases (from a mean value
of –I in pyrite to as high as VI for the predominant dissolution product sulphate SO42–), decreases
(from –I to –II), or remains unchanged in the case of oxidative, reductive, and chemical dissolution,
respectively. Upon dissolution, the oxidation state of the Fe remains unchanged at II (unless it is
subsequently oxidised by O2 under oxic conditions).
There is an extensive literature on the oxidative dissolution of FeS2p because of the importance of
this process in acid rock drainage and mineral extraction. The predominant product of the reaction
is SO42–, although smaller amounts of intermediate oxidation products such as sulphur, sulphite,
thiosulphate, etc. are also reported.
There has been far less study of the reductive dissolution of pyrite, a process that not only leads
to the release of sulphide but also produces less-stable (and, hence, more-soluble) iron sulphides
such as pyrrhotite or troilite. The reduction of pyrite is well known in H2 gas atmospheres at tem
peratures above 400°C, and there is also evidence for the reaction in aqueous solution at temperatures
of 90–180°C with H2 overpressures of 0.8 MPa and 1.8 MPa. There is no evidence for microbially
enhanced pyrite reductive dissolution.
The chemical dissolution of pyrite has been demonstrated in acid solution, but no detailed study of
the dissolution kinetics has been published.
In terms of the probable behaviour of pyrite in the repository, the predominant process will be the
oxidative dissolution of FeS2p during the early warm, aerobic period. This will result in the consumption
of a portion of the initially trapped atmospheric O2 and the formation of SO42– and, possibly, smaller
quantities of other sulphur and sulphur oxyanions species, some of which may support corrosion of
the canister if they reach the copper surface.
During the long-term anaerobic phase, oxidation of pyrite in the backfill may continue if there is a
suitable electron acceptor (such as nitrate NO3–) available and suitable conditions for microbial activity.
Because of the use of highly compacted bentonite, there will be no microbial activity and no oxidative
dissolution of pyrite in the buffer during the anaerobic period. It seems unlikely that either the tem
perature and/or H2 pressure will be high enough in the repository to sustain the reductive dissolution
of pyrite during this period. Therefore, only chemical dissolution of pyrite is likely in the long term,
with the rate limited by the slow diffusion of HS–/HS2– away from the dissolving surface.
SKB TR-13-1945
In summary, therefore, the pyrite in the buffer and backfill material is likely to consume a fraction
of the initially trapped atmospheric O2 but is unlikely to produce a significant quantity of dissolved
sulphur species that could support corrosion of the canister.
46
SKB TR-13-19
References
SKB’s (Svensk Kärnbränslehantering AB) publications can be found at www.skb.se/publications.
Abraitis P K, Pattrick R A D, Vaughan D J, 2004. Variations in the compositional, textural and
electrical properties of natural pyrite: a review. International Journal of Mineral Processing 74,
41–59.
Ahlberg E, Broo A E, 1996. Oxygen reduction at sulphide minerals. 3. The effect of surface pretreatment on the oxygen reduction at pyrite. International Journal of Mineral Processing 47, 49–60.
Ahlberg E, Forssberg K S E, Wang X, 1990. The surface oxidation of pyrite in alkaline solution.
Journal of Applied Electrochemistry 20, 1033–1039.
Arcos D, Grandia F, Domènech C, Fernández A M, Villar M V, Muurinen A, Carlsson T,
Sellin P, Hernán P, 2008. Long-term geochemical evolution of the near field repository: insights
from reactive transport modelling and experimental evidences. Journal of Contaminant Hydrology
102, 196–209.
Benning L G, Wilkin R T, Barnes H L, 2000. Reaction pathways in the Fe–S system below 100°C.
Chemical Geology 167, 25–51.
Berner R A, 1967. Thermodynamic stability of sedimentary iron sulfides. American Journal of
Science 265, 773–785.
Betelu S, Lerouge C, Berger G, Giffaut E, Ignatiadis I, 2012. Mechanistic study of pyrite reduction
by hydrogen in NaCl 0.1 M at 90°C using electrochemical techniques. In 5th International meeting
“Clays in natural and engineered barriers for radioactive waste confinement”, Montpellier, France,
22–25 October 2012. Book of abstracts.
Biegler T, 1976. Oxygen reduction on sulphide minerals. Part II. Relation between activity and
semiconducting properties of pyrite electrodes. J. Electroanal. Chem. 70, 265–275.
Biegler T, Swift D A, 1979. Anodic behaviour of pyrite in acid solutions. Electrochimica Acta 24,
415–420.
Bildstein O, Trotignon L, Perronnet M, Jullien M, 2006. Modelling iron–clay interactions in deep
geological disposal conditions. Physics and Chemistry of the Earth 31, 618–625.
Borda M J, Strongin D R, Schoonen M A, 2003. A vibrational spectroscopic study of the oxidation
of pyrite by ferric ion. Amer. Mineral. 88, 1318–1323.
Borda M J, Strongin D R, Schoonen M A, 2004. A vibrational spectroscopic study of the oxidation
of pyrite by molecular oxygen. Geochimica et Cosmochimica Acta 68, 1807–1813.
Bosch J, Meckenstock R U, 2012. Rates and potential mechanism of anaerobic nitrate-dependent
microbial pyrite oxidation. Biochemical Society Transactions 40, 1280–1283.
Bosch J, Lee K-Y, Jordan G, Kim K-W, Meckenstock R U, 2012. Anaerobic, nitrate-dependent
oxidation of pyrite nanoparticles by Thiobacillus denitrificans. Environmental Science & Technology
46, 2095–2101.
Bruggeman C, Maes A, Vancluysen J, 2007. The identification of FeS2 as a sorption sink for
Tc(IV). Physics and Chemistry of the Earth 32, 573–580.
Brunner B, Yu J-Y, Mielke R E, MacAskill J A, Madzunkov S, McGenity T J, Coleman M,
2008. Different isotope and chemical patterns of pyrite oxidation related to lag and exponential
growth phases of Acidothiobacillus ferrooxidans reveal a microbial growth strategy. Earth and
Planetary Science Letters 270, 63–72.
Butler I B, Rickard D, 2000. Framboidal pyrite formation via the oxidation of iron (II) monosulfide
by hydrogen sulphide. Geochimica et Cosmochimica Acta 64, 2665–2672.
Cabral T, Ignatiadis I, 2001. Mechanistic study of the pyrite–solution interface during the oxidative
bacterial dissolution of pyrite (FeS2) by using electrochemical techniques. International Journal of
Mineral Processing 62, 41–64.
SKB TR-13-1947
Cai Y, Pan Y, Xue J, Sun Q, Su G, Li X, 2009. Comparative XPS study between experimentally
and naturally weathered pyrites. Applied Surface Science 255, 8750–8760.
Caldeira C L, Ciminelli V S T, Dias A, Osseo-Asare K, 2003. Pyrite oxidation in alkaline solutions:
nature of the product layer. International Journal of Mineral Processing 72, 373–386.
Caldeira C L, Ciminelli V S T, Osseo-Asare K, 2010. The role of carbonate ions in pyrite oxidation
in aqueous systems. Geochimica et Cosmochimica Acta 74, 1777–1789.
Chandra A P, Gerson A R, 2010. The mechanisms of pyrite oxidation and leaching: a fundamental
perspective. Surface Science Reports 65, 293–315.
Chase M W (ed), 1985. JANAF thermochemical tables. 3rd ed. Part I, Al-Co; Part II, Cr-Zr. (Journal
of Physical and Chemical Reference Data 14, Supplement 1)
Chase M W (ed), 1998. NIST-JANAF thermochemical tables. 4th ed. (Journal of Physical and
Chemical Reference Data, Monograph 9)
Christidis G E, Huff W D, 2009. Geological aspects and genesis of bentonites. Elements 5, 93–98.
Ciminelli V S T, Osseo-Asare K, 1995. Kinetics of pyrite oxidation in sodium carbonate solutions.
Metallurgical and Materials Transactions B 26, 209–218.
CRC, 1982. Handbook of chemistry and physics: a ready-reference book of chemical and physical
data. 63rd ed. 1982–1983. Boca Raton, FL: CRC Press.
Cruz R, Bertrand V, Monroy M, González I, 2001. Effect of sulfide impurities on the reactivity
of pyrite and pyritic concentrates: a multi-tool approach. Applied Geochemistry 16, 803–819.
Davison W, 1991. The solubility of iron sulphides in synthetic and natural waters at ambient tem
perature. Aquatic Sciences 53, 309–329.
de Donato P, Mustin C, Benoit R, Erre R, 1993. Spatial distribution of iron and sulphur species
on the surface of pyrite. Applied Surface Science 68, 81–93.
Delécaut G, 2004. The geochemical behaviour of uranium in the Boom Clay. PhD thesis. Université
catholique de Louvain, Belgium.
Descostes M, Mercier F, Beaucaire C, Zuddas P, Trocellier P, 2001. Nature and distribution
of chemical species on oxidized pyrite surface: complementarity of XPS and nuclear microprobe
analysis. Nuclear Instruments and Methods in Physics Research Section B 181, 603–609.
Descostes M, Beaucaire C, Mercier F, Savoye S, Sow J, Zuddas P, 2002. Effect of carbonate ions
on pyrite (FeS2) dissolution. Bulletin de la Societe Geologique de France 173, 265–270.
Descostes M, Vitorge P, Beaucaire C, 2004. Pyrite dissolution in acidic media. Geochimica et
Cosmochimica Acta 68, 4559–4569.
Descostes M, Schlegel M L, Eglizaud N, Descamps F, Miserque F, Simoni E, 2010. Uptake of
uranium and trace elements in pyrite (FeS2) suspensions. Geochimica et Cosmochimica Acta 74,
1551–1562.
Druschel G, Borda M, 2006. Comment on “Pyrite dissolution in acidic media” by M. Descostes, P.
Vitorge, and C. Beaucaire. Geochimica et Cosmochimica Acta 70, 5246–5250.
Elsetinow A R, Guevremont J M, Strongin D R, Schoonen M A A, Strongin M, 2000. Oxidation
of {100} and {111} surfaces of pyrite: effects of preparation method. American Mineralogist 85,
623–626.
Fernández A M, Baeyens B, Bradbury M, Rivas P, 2004. Analysis of the porewater chemical
composition of a Spanish compacted bentonite used in an engineered barrier. Physics and Chemistry
of the Earth 29, 105–118.
Fukushi K, Sugiura T, Morishita T, Takahashi Y, Hasebe N, Ito H, 2010. Iron–bentonite inter
actions in the Kawasaki bentonite deposit, Zao area, Japan. Applied Geochemistry 25, 1120–1132.
Giannetti B F, Almeida C M V B, Bonilla S H, 2006. Electrochemical kinetic study of surface
layer growth on natural pyrite in acid medium. Colloids and Surfaces A: Physicochemical and
Engineering Aspects 272, 130–138.
48
SKB TR-13-19
Gleisner M, Herbert R B, Frogner Kockum P C, 2006. Pyrite oxidation by Acidothiobacillus
ferrooxidans at various concentrations of dissolved oxygen. Chemical Geology 225, 16–29.
Grandia F, Domènech C, Arcos D, Duro L, 2006. Assessment of the oxygen consumption in
the backfill. Geochemical modelling in a saturated backfill. SKB R-06-106, Svensk Kärnbränsle
hantering AB.
Grønvold F, Westrum E F, 1976. Heat capacities of iron disulfides from 5 to 700 K, pyrite from
300 to 780 K, and the transformation of marcasite to pyrite. The Journal of Chemical Thermo
dynamics 8, 1039–1048.
Guevremont J M, Elsetinow A R, Strongin D R, Bebie J, Schoonen M A A, 1998a. Structure
sensitivity of pyrite oxidation: comparison of the (100) and (111) planes. American Mineralogist 83,
1353–1356.
Guevremont J M, Bebie J, Elsetinow A R, Strongin D R, Schoonen M A A, 1998b. Reactivity
of the (100) plane of pyrite in oxidizing gaseous and aqueous environments: effects of surface imper
fections. Environmental Science & Technology 32, 3743–3748.
Hamilton I C, Woods R, 1981. An investigation of surface oxidation of pyrite and pyrrhotite by
linear potential sweep voltammetry. Journal of Electroanalytical Chemistry and Interfacial Electro
chemistry 118, 327–343.
Heidel C, Tichomirowa, 2010. The role of dissolved molecular oxygen in abiotic pyrite oxidation
under acid pH conditions – experiments with 18O‑enriched molecular oxygen. Applied Geochemistry
25, 1664–1675.
Hol A, van der Weijden R D, Van Weert G, Kondos P, Buisman C J N, 2010. Bio-reduction of
pyrite investigated in a gas lift loop reactor. International Journal of Mineral Processing 94, 140–146.
Holmes P R, Crundwell F K, 2000. The kinetics of the oxidation of pyrite by ferric ions and dis
solved oxygen: an electrochemical study. Geochimica et Cosmochimica Acta 64, 263–274.
Holmes P R, Fowler T A, Crundwell F K, 1999. The mechanism of bacterial action in the leaching
of pyrite by Thiobacillus ferrooxidans. An electrochemical study. Journal of The Electrochemical
Society 146, 2906–2912.
Huminicki D M C, Rimstidt J D, 2009. Iron oxyhydroxide coating of pyrite for acid mine drainage
control. Applied Geochemistry 24, 1626–1634.
Hummel W, Berner U, Curti E, Pearson F J, Thoenen T, 2002. Nagra/PSI chemical thermo
dynamic data base 01/01. Parkland, FL: Universal Publishers, Section 5.6.3.
Jerz J K, Rimstidt J D, 2004. Pyrite oxidation in moist air. Geochimica et Cosmochimica Acta 68,
701–714.
Joeckel R M, Ang Clement B J, VanFleet Bates L R, 2005. Sulfate-mineral crusts from pyrite
weathering and acid rock drainage in the Dakota Formation and Graneros Shale, Jefferson County,
Nebraska. Chemical Geology 215, 433–452.
Jørgensen C J, Jacobsen O S, Elberling B, Aamand J, 2009. Microbial oxidation of pyrite coupled
to nitrate reduction in anoxic groundwater sediment. Environmental Science & Technology 43,
4851–4857.
Kamyshny A, Goifman A, Gun, J, Rizkov D, Lev O, 2004. Equilibrium distribution of polysulfide
ions in aqueous solutions at 25°C: a new approach for the study of polysulfides’ equilibria. Environ
mental Science & Technology 38, 6633–6644.
Karnland O, 2010. Chemical and mineralogical characterization of the bentonite buffer for the
acceptance control procedure in a KBS-3 repository. SKB TR-10-60, Svensk Kärnbränslehantering AB.
Karnland O, Olsson S, Nilsson U, 2006. Mineralogy and sealing properties of various bentonites
and smectite-rich clay materials. SKB TR-06-30, Svensk Kärnbränslehantering AB.
Kaye G W C, Laby T H, 1986. Table of physical and chemical constants. 15th ed. London:
Longman, Section 2.10.5.
SKB TR-13-1949
Kaye & Laby, 2005. Tables of physical & chemical constants (16th edition 1995), Version 1.0.
Available at http://www.kayelaby.npl.co.uk/ [25 October 2012].
Kelsall G H, Yin Q, Vaughan D J, England K E R, Brandon N P, 1999. Electrochemical oxidation
of pyrite (FeS2) in aqueous electrolytes. Journal of Electroanalytical Chemistry 471, 116–125.
King F, 2012. Gaseous hydrogen issues in nuclear waste disposal. In Gangloff R P, Somerday B P
(eds). Gaseous hydrogen embrittlement of materials in energy technologies. Volume 1: the problem, its
characterisation and effects on particular alloy classes. Cambridge: Woodhead Publishing, 126–148.
King F, Lilja C, Pedersen K, Pitkänen P, Vähänen M, 2010. An update of the state-of-the-art report
on the corrosion of copper under expected conditions in a deep geologic repository. SKB TR-10-67,
Svensk Kärnbränslehantering AB.
King F, Kolar M, Vähänen M, 2011a. Reactive-transport modelling of the sulphide-assisted cor
rosion of copper nuclear waste canisters. In Kursten B, Feron D, Druyts F (eds). Sulphur-assisted
corrosion in nuclear disposal systems. Leeds, UK: Maney Publishing. (European Federation of
Corrosion Publications 59), 152–164.
King F, Kolar M, Vähänen M, Lilja C, 2011b. Modelling long term corrosion behaviour of copper
canisters in KBS-3 repository. Corrosion Engineering, Science and Technology 46, 217–222.
Kolaříková I, Švandová J, Přikryl R, Vinšová H, Jedináková-Křížová V, Zeman J, 2010.
Mineralogical changes in bentonite barrier within Mock-Up-CZ experiment. Applied Clay Science
47, 10–15.
Laajalehto K, Leppinen J, Kartio I, Laiho T, 1999. XPS and FTIR study of the influence of elec
trode potential on activation of pyrite by copper or lead. Colloids and Surfaces A: Physicochemical
and Engineering Aspects 154, 193–199.
Lambert J M, Simkovich G, Walker P L, 1980. Production of pyrrhotites by pyrite reduction.
Fuel 59, 687–690.
Lambert J M, Simkovich G, Walker P L, 1998. The kinetics and mechanism of the pyrite-to-pyrrhotite
transformation. Metallurgical and Materials Transactions B 29, 385–396.
Lange K, Rowe R K, Jamieson H, Flemming R L, Lanzirotti A, 2010. Characterization of geo
synthetic clay liner bentonite using micro-analytical methods. Applied Geochemistry 25, 1056–1069.
Langmuir D, 1997. Aqueous environmental geochemistry. Upper Saddle River, NJ: Prentice Hall.
Lazo C, Karnland O, Tullborg E-L, Puigdomenech I, 2003. Redox properties of MX-80 and
Montigel bentonite-water systems. In Finch R J, Bullen D B (eds). Scientific basis for nuclear waste
management XXVI. Warrendale, PA: Materials Research Society. (Materials Research Society
Symposium Proceedings 757), paper II8.1.
Lefticariu L, Pratt L A, LaVerne J A, Schimmelmann A, 2010. Anoxic pyrite oxidation by water
radiolysis products – a potential source of biosustaining energy. Earth and Planetary Science Letters
292, 57–67.
Lehner S, Savage K, Ciobanu M, Cliffel D E, 2007. The effect of As, Co, and Ni impurities on
pyrite oxidation kinetics: an electrochemical study of synthetic pyrite. Geochimica et Cosmochimica
Acta 71, 2491–2509.
Licht S, Davis J, 1997. Disproportionation of aqueous sulphur and sulfide: kinetics of polysulfide
decomposition. Journal of Physical Chemistry B 101, 2540–2545.
Liu R, Wolfe A L, Dzombak D A, Horwitz C P, Stewart B W, Capo R C, 2008. Electrochemical
study of hydrothermal and sedimentary pyrite dissolution. Applied Geochemistry 23, 2724–2734.
Liu R, Wolfe A L, Dzombak D A, Horwitz C P, Stewart B W, Capo R C, 2009. Controlled electro
chemical dissolution of hydrothermal and sedimentary pyrite. Applied Geochemistry 24, 836–842.
Lizama H M, Suzuki I, 1989. Rate equations and kinetic parameters of the reactions involved
in pyrite oxidation by Thiobacillus ferrooxidans. Applied and Environmental Microbiology 55,
2918–2923.
50
SKB TR-13-19
Luther G W, 1987. Pyrite oxidation and reduction: molecular orbital theory consideration. Geochimica
et Cosmochimica Acta 51, 3193–3199.
Macdonald D D, Sharifi-Asl S, 2011. Is copper immune to corrosion when in contact with water
and aqueous solutions? Research Report 2011:09, Strålsäkerhetsmyndigheten (Swedish Radiation
Safety Authority).
Marty N C M, Fritz B, Clément A, Michau N, 2010. Modelling the long term alteration of the
engineered bentonite barrier in an underground radioactive waste repository. Applied Clay Science 47,
82–90.
Melamed A, Pitkänen O, 1994. Water-compacted Na-bentonite interaction in simulated nuclear
fuel disposal conditions: the role of accessory minerals. In Barkatt A, Van Konynenburg R A (eds).
Scientific basis for nuclear waste management XVII: symposium held in Boston, Massachusetts,
USA, 29 November – 3 December. Pittsburgh, PA: Materials Research Society. (Materials Research
Society Symposium Proceedings 333), 919–924.
Melamed A, Pitkänen O, 1996. Chemical and mineralogical aspects of water–bentonite interaction in
nuclear fuel disposal conditions. VTT Research Notes 1766, Technical Research Centre of Finland.
Montes-H G, Fritz B, Clement A, Michau N, 2005a. Modeling of transport and reaction in an
engineered barrier for radioactive waste confinement. Applied Clay Science 29, 155–171.
Montes-H G, Marty N, Fritz B, Clement A, Michau N, 2005b. Modelling of long-term diffusionreaction in a bentonite barrier for radioactive waste confinement. Applied Clay Science 30, 181–198.
Montes-H G, Fritz B, Clement A, Michau N, 2005c. Modelling of geochemical reactions and
experimental cation exchange in MX 80 bentonite. Journal of Environmental Management 77, 35–46.
Moses C O, Herman J S, 1991. Pyrite oxidation at circumneutral pH. Geochimica et Cosmochimica
Acta 55, 471–482.
Moses C O, Nordstrom D K, Herman J S, Mills A L, 1987. Aqueous pyrite oxidation by dissolved
oxygen and by ferric iron. Geochimica et Cosmochimica Acta 51, 1561–1571.
Moslemi H, Shamsi P, Habashi F, 2011. Pyrite and pyrrhotite open circuit potentials study: effects
on flotation. Minerals Engineering 24, 1038–1045.
Murphy R, Strongin D R, 2009. Surface reactivity of pyrite and related sulfides. Surface Science
Reports 64, 1–45.
Mycroft J R, Bancroft G M, McIntyre N S, Lorimer J W, Hill I R, 1990. Detection of sulphur
and polysulphides on electrochemically oxidized pyrite surfaces by X‑ray photoelectron spectro
scopy and Raman spectroscopy. Journal of Electroanalytical Chemistry and Interfacial Electro
chemistry 292, 139–152.
Nagra, 2002. Project Opalinus Clay. Safety report. Demonstration of disposal feasibility of spent
fuel, vitrified high-level waste and long-lived intermediate-level waste (Entsorgungsnachweis).
Nagra Technical Report NTB 02-05, Nagra, Switzerland.
Naveau A, Monteil-Rivera F, Guillon E, Dumonceau J, 2006. XPS and XAS studies of copper(II)
sorbed onto a synthetic pyrite surface. Journal of Colloid and Interface Science 303, 25–31.
Nesbitt H W, Bancroft G M, Pratt A R, Scaini M J, 1998. Sulfur and iron surface states on fractured
pyrite surfaces. American Mineralogist 83, 1067–1076.
Nicholson R V, Gillham R W, Reardon E J, 1988. Pyrite oxidation in carbonate-buffered solution:
1. Experimental kinetics. Geochimica et Cosmochimica Acta 52, 1077–1085.
Nicholson R V, Gillham R W, Reardon E J, 1990. Pyrite oxidation in carbonate-buffered solution:
21. Rate control by oxide coatings. Geochimica et Cosmochimica Acta 54, 395–402.
Ochs M, Lothenbach B, Shibata M, Yui M, 2004. Thermodynamic modelling and sensitivity analysis
of porewater chemistry in compacted bentonite. Physics and Chemistry of the Earth 29, 129–136.
Ohkubo T, Kikuchi H, Yamaguchi M, 2008. An approach of NMR relaxometry for understanding
water in saturated compacted bentonite. Physics and Chemistry of the Earth 33, S169–S176.
SKB TR-13-1951
Ohmoto H, Hayashi K-I, Kajisa Y, 1994. Experimental study of the solubilities of pyrite in NaClbearing aqueous solutions at 250–350°C. Geochimica et Cosmochimica Acta 58, 2169–2185.
Puigdomenech I, Trotignon L, Kotelnikova S, Pedersen K, Griffault L, Michaud V, Lartigue J-E,
Hama K, Yoshida H, West J M, Bateman K, Milodowski A, Banwart S A, Rivas Perez J,
Tullborg E-L, 2000. O2 consumption in a granitic environment. In Smith R W, Shoesmith D W (eds).
Scientific basis for nuclear waste management XXIII: symposium held in Boston, Massachusetts,
USA, 29 November – 2 December 1999. Warrendale, PA: Materials Research Society. (Materials
Research Society Symposium Proceedings 608), 179–184.
Puigdomenech I, Ambrosi J-P, Eisenlohr L, Lartigue J-E, Banwart S A, Bateman K,
Milodowski A E, West J M, Griffault L, Gustafsson E, Hama K, Yoshida H, Kotelnikova S,
Pedersen K, Michaud V, Trotignon L, Rivas Perez J, Tullborg E-L, 2001. O2 depletion in granitic
media. The REX project. SKB TR-01-05, Svensk Kärnbränslehantering AB.
Pusch R, 2001. The buffer and backfill handbook. Part 2: materials and techniques. SKB TR-02-12,
Pusch R, 2003. The buffer and backfill handbook. Part 3: models for calculation of processes and
behaviour. SKB TR-03-07, Svensk Kärnbränslehantering AB.
Rand D A J, 1977. Oxygen reduction on sulphide minerals. Part III. Comparison of activities of
various copper, iron, lead and nickel mineral electrodes. Journal of Electroanalytical Chemistry and
Interfacial Electrochemistry 83, 19–32.
Rawlings D E, Tributsch H, Hansford G S, 1999. Reasons why ‘Leptospirillum’-like species
rather than Thiobacillus ferrooxidans are the dominant iron-oxidizing bacteria in many commercial
processes for the biooxidation of pyrite and related ores. Microbiology 145, 5–13.
Rickard D, 2006. The solubility of FeS. Geochimica et Cosmochimica Acta 70, 5779–5789.
Rickard D, Luther G W, 2007. Chemistry of iron sulfides. Chemical Reviews 107, 514–562.
Rimstidt J D, Vaughan D J, 2003. Pyrite oxidation: a state-of-the-art assessment of the reaction
mechanism. Geochimica et Cosmochimica Acta 67, 873–880.
Robie R A, Hemingway B S, 1995. Thermodynamic properties of minerals and related substances
at 298.15 K and 1 bar (105 pascals) pressure and at higher temperatures. U.S. Geological Survey
Bulletin 2131.
Robie R A, Hemingway B S, Fisher J R, 1978. Thermodynamic properties of minerals and related
substances at 298.15 K and 1 bar (105 pascals) pressure and at higher temperatures. U.S. Geological
Survey Bulletin 1452.
Savage K S, Stefan D, Lehner S W, 2008. Impurities and heterogeneity in pyrite: influences on
electrical properties and oxidation products. Applied Geochemistry 23, 103–120.
Schippers A, Jørgensen B B, 2001. Oxidation of pyrite and iron sulfide by manganese dioxide in
marine sediments. Geochimica et Cosmochimica Acta 65, 915–922.
Schippers A, Jørgensen B B, 2002. Biogeochemistry of pyrite and iron sulfide oxidation in marine
sediments. Geochimica et Cosmochimica Acta 66, 85–92.
Schoonen, M A A, 2004. Mechanisms of sedimentary pyrite formation. In Amend J P, Edwards K J,
Lyons T W (eds). Sulfur biogeochemistry – past an present. Boulder, CO: Geological Society of
America. (Geological Society of America Special Paper 379), 117–134.
Schoonen M, Elsetinow A, Borda M, Strongin D, 2000. Effect of temperature and illumination on
pyrite oxidation between pH 2 and 6. Geochemical Transactions 4. doi:10.1039/b004044o.
Schoonen M A A, Harrington A D, Laffers R, Strongin D R, 2010. Role of hydrogen peroxide and
hydroxyl radical in pyrite oxidation by molecular oxygen. Geochimica Cosmochimica et Acta 74,
4971–4987.
Scott T B, Riba Tort O, Allen G C, 2007. Aqueous uptake of uranium onto pyrite surfaces; reactivity
of fresh versus weathered material. Geochimica Cosmochimica et Acta 71, 5044–5053.
52
SKB TR-13-19
Sidborn M, Neretnieks I, 2003. Modelling of biochemical processes in rocks: oxygen depletion by
pyrite oxidation – model development and exploratory simulations. In Finch R J, Bullen D B (eds).
Scientific basis for nuclear waste management XXVI. Warrendale, PA: Materials Research Society.
(Materials Research Society Symposium Proceedings 757), paper II11.5.
Singh A K, Pourbaix A, 1997. E-pH diagrams for the Fe-S-H2O system from 25 to 150°C. Cebelcor
Report RT.318, Centre Belge d’Etude de la Corrosion.
SKB, 2010a. Corrosion calculations report for the safety assessment SR-Site. SKB TR-10-66,
SKB, 2010b. Design, production and initial state of the buffer. SKB TR-10-15, Svensk Kärnbränsle
hantering AB.
SKB, 2011. Long-term safety for the final repository for spent nuclear fuel at Forsmark. Main report
of the SR-Site project. Volume I. SKB TR-11-01, Svensk Kärnbränslehantering AB.
Szakálos P, Hultquist G, Wikmark G, 2007. Corrosion of copper by water. Electrochemical and
Solid-State Letters 10, C63–C67.
Tao, D P, Li Y Q, Richardson P E, Yoon R‑H, 1994. The incipient oxidation of pyrite. Colloids and
Surfaces A: Physicochemical and Engineering Aspects 93, 229–239.
Tao D P, Richardson P E, Luttrell G H, Yoon R‑H, 2003. Electrochemical studies of pyrite
oxidation and reduction using freshly-fractured electrodes and rotating ring-disc electrodes.
Electrochimica Acta 48, 3615–3623.
Taylor B E, Wheeler M C, Nordstrom D K, 1984. Stable isotope geochemistry of acid mine
drainage: experimental oxidation of pyrite. Geochimica et Cosmochimica Acta 48, 2669–2678.
Techer I, Clauer N, Liewig N, 2009. Ageing effect on the mineral and chemical composition
of Opalinus Clays (Mont Terri, Switzerland) after excavation and surface storage. Applied
Geochemistry 24, 2000–2014.
Thomas J E, Skinner W M, Smart R St C, 2001. A mechanism to explain sudden changes in rates
and products for pyrrhotite dissolution in acid solution. Geochimica et Cosmochimica Acta 65, 1–12.
Todd E C, Sherman D M, Purton J A, 2003. Surface oxidation of pyrite under ambient atmospheric
and aqueous (pH=2 to 10) conditions: electronic structure and mineralogy from X‑ray absorption
spectroscopy. Geochimica et Cosmochimica Acta 67, 881–893.
Toniazzo V, Mustin C, Portal J M, Humbert B, Benoit, R, Erre R, 1999. Elemental sulfur at the
pyrite surfaces: speciation and quantification. Applied Surface Science 143, 229–237.
Torrentó C, Cama J, Urmeneta J, Otero N, Soler A, 2010. Denitrification of groundwater with
pyrite and Thiobacillus denitrificans. Chemical Geology 278, 80–91.
Toulmin P T, Barton P B, 1964. A thermodynamic study of pyrite and pyrrhotite. Geochimica et
Cosmochimica Acta 28, 641–671.
Truche L, Berger G, Destrigneville C, Guillaume D, Giffaut E, 2010. Kinetics of pyrite to
pyrrhotite reduction by hydrogen in calcite buffered solutions between 90 and 180°C: implications
for nuclear waste disposal. Geochimica et Cosmochimica Acta 74, 2894–2914.
Usher C R, Cleveland C A, Strongin D R, Schoonen M A, 2004. Origin of oxygen in sulfate
during pyrite oxidation with water and dissolved oxygen: an in situ horizontal attenuated total
reflectance spectroscopy isotope study. Environmental Science & Technology 38, 5604–5606.
Van Loon L R, Glaus M A, Müller W, 2007. Anion exclusion effects in compacted bentonites:
towards a better understanding of anion diffusion. Applied Geochemistry 22, 2536–2552.
Vaughan D J, 2005. Minerals/Sulphides. In Encyclopedia of geology. Amsterdam: Elsevier,
574–586.
von Oertzen G U, Skinner W M, Nesbitt H W, Pratt A R, Buckley A N, 2007. Cu adsorption on
pyrite (100): ab initio and spectroscopic studies. Surface Science 601, 5794–5799.
SKB TR-13-1953
Wagman D D, Evans W H, Parker V B, Schumm R H, Halow I, Bailey S M, Churney K L,
Nuttall R L, 1982. The NBS tables of chemical thermodynamic properties: selected values for
inorganic and C1 and C2 organic substances in SI units. New York: American Chemical Society.
(Journal of Physical and Chemical Reference Data 11, Supplement 2)
Weerasooriya R, Tobschall H J, 2005. Pyrite–water interactions: effects of pH and pFe on surface
charge. Colloids and Surfaces A: Physicochemical and Engineering Aspects 264, 68–74.
Weisener C, Gerson A, 2000. An investigation of the Cu(II) adsorption mechanism on pyrite by
ARXPS and SIMS. Minerals Engineering 13, 1329–1340.
Wersin P, Spahiu K, Bruno J, 1994. Time evolution of dissolved oxygen and redox conditions in
a HLW repository. SKB TR 94-02, Svensk Kärnbränslehantering AB.
Williamson M A, Rimstidt J D, 1994. The kinetics and electrochemical rate-determining step of
aqueous pyrite oxidation. Geochimica et Cosmochimica Acta 58, 5443–5454.
Zhang G, Samper J, Montenegro L, 2008. Coupled thermo-hydro-bio-geochemical reactive trans
port model of CERBERUS heating and radiation experiment in Boom clay. Applied Geochemistry
23, 932–949.
Zhao W, Zhu H, Zong Z-M, Xia J-H, Wei X-Y, 2005. Electrochemical reduction of pyrite in aqueous
NaCl solution. Fuel 84, 235–238.
Zhao W, Xu W‑J, Zhong S‑T, Zong Z‑M, 2008. Desulfurization of coal by an electrochemicalreduction flotation technique. Journal of China University of Mining and Technology 18, 571–574.
54
SKB TR-13-19

A review of the properties of pyrite and the implications for

Transcription

Similar documents

PDF Report - Search ARIS Database

Vein Style Deposits

1996-15 - Ministry of Energy and Mines and Responsible for Core

chapter 8 paragenesis, formation conditions, and fluid inclusion

PDF Report

PDF Report

problems connected with the use of iron metal in reactive barrier

Nanotechnology in oral dosage forms. A. Sallam. Consultant. TQ

ibostar plus. - Daetwyler USA